Find the best AI models you can run
See the code
The open AI model leaderboard. Every model is ranked by one score built from 19,731 published benchmark results, with its price, the memory it needs and the command to run it.
Filter by your computer or budget at whatmodel.sh, or read the same leaderboard below. About this repo
llm · image · video · tts · stt
updated 2026-10-11 02:20 UTC · 830 models · scores combine 472 benchmarks · how it works · benchmark your machine · interactive version
Run curl "whatmodel.sh/?mac=$(sysctl -n hw.memsize)" to see what your Mac can run, or curl whatmodel.sh for this list in your terminal.
curl "whatmodel.sh/?ram=$(awk '/MemT/{print $2}' /proc/meminfo)&vram=$(nvidia-smi -i 0 --query-gpu=memory.total --format=csv,noheader,nounits 2>/dev/null || cat /sys/class/drm/card*/device/mem_info_vram_total 2>/dev/null | sort -n | tail -1)" # Linux
curl.exe "whatmodel.sh/?ram=$((gcim Win32_ComputerSystem).TotalPhysicalMemory)&vram=$(nvidia-smi -i 0 --query-gpu=memory.total --format=csv,noheader,nounits 2>$null)" # Windows
16GB Mac · 32GB Mac · 64GB Mac · 128GB Mac · 12GB graphics card · 24GB graphics card · 16GB, no graphics card
score vs API price
│ MiMo-V2.6-Pro GPT-6.1 Sol ◆──────◆────────◆ GPT-6 Astra
75 ┤ · · · · · · · · ●─●──∘─●◆∘∘●∘───◇∘●─∘─∘◇∘∘∘∘∘∘◇◇∘──◇◇◇◇◇◇◇◇ ·◇·◇◇ · · ◇ · ·◇·◇· ◇ ◇ ·
│ ◆─∘──∘∘∘∘∘∘ ◇∘∘ ◇∘∘∘∘∘ ∘◇∘∘∘◇∘∘∘ ∘◇◇ ∘ ∘◇◇∘◇◇ ◇ ◇ ◇
50 ┤ · · · · · ●┘◇∘∘ ·∘◇∘∘ · ·∘·∘∘ · · · ·◇· · ·◇· · · · · ·∘· · · ··· · · · · · ◇ · · · ·
│ ●────────┘
└──────────────────┬──────────────────────┬──────────────────────┬─────────────────────
$0.1 $1 $10
score vs API price (USD per 1M tokens, 3 input : 1 output, log)
● open weights ◆ API only ─┘ best for the price
│ Claude Opus 5.5
│ · · ◆────────◆ GPT-6 Astra
│ GPT-6.1 Sol ◆──────┘ ◇ ◇
80 ┤ · · · · · · · · ··· MiMo-V2.6-Pro ●───────∘───────────┘ · · ·◇··◇ · · ◇ · · · · · ◇ ·
│ ●◆─◇●────◇─┘ ◇∘ ◇∘ ◇ ◇ ◇∘ ◇ ◇
│ ●─●────┘ ∘ ◇ ·◇ ∘◇◇ ∘ ◇◇◇◇ ◇ ◇ ◇ ◇
│ │ ∘ ∘ ∘∘ ◇∘∘ ◇ ∘◇∘∘∘∘∘∘◇◇∘ ◇ ◇ ◇
│ ◆────┘· ∘ ◇∘ ∘ ∘ ∘ ∘∘∘ ∘∘∘ ◇◇ ◇ ∘◇ · ◇
60 ┤ · · · · · ·│· · ∘ ∘ ◇∘· · ·∘∘∘·∘∘ · ◇∘◇∘∘ ∘ · · · · ◇◇∘ · · · · · · · · · · · · · · ·
│ ● ∘ ∘·∘∘∘ ◇ ∘ ∘ ∘ ∘ ◇∘∘∘◇· ◇ ◇ ·
│ │ ∘∘∘ ∘ ◇∘ ◇∘ ∘ ∘ ∘ ∘ ◇ ◇
│ │ ∘ ·◇ ∘ ∘ · ◇ · ◇
│ ●┘◇ ∘ ∘ ∘ ∘ ◇ ∘
│ │ ∘ · ∘ · ·
40 ┤ · · · · · │ · · · ·∘· · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · ·
│ │ · · ·
│ ●─Llama 3.2 1B Instruct
│ · · ·
└──────────────────┬──────────────────────┬──────────────────────┬─────────────────────
$0.1 $1 $10
score vs memory to run it (GB, log)
─┘ best for the memory
│ Hy4 preview (770B-A49B)
│ · MiMo-V2.6-Flash (309B-A15… ●────∘● MiMo-V2.6-Pro
75 ┤ · · · · · · · ·Qwen3.8 27B·●────────────●────∘∘●─∘●─────∘┘· ∘∘· · ∘∘· ·∘·∘· · · · ∘ ·
│ Spark-X2.5 4B ∘∘∘ ∘ ∘∘ ∘ ∘∘∘∘ ∘∘∘ ∘ ∘∘ ∘ ∘∘ · ∘
│ ●──●────∘─∘∘─∘∘─────∘∘∘∘∘∘ ∘ ∘ ∘∘∘ ∘∘∘ ∘∘ ∘∘ ∘∘ ∘ ∘
│ ∘ ∘ ∘ ∘∘ ∘∘ ∘∘∘∘∘∘∘ ∘∘∘∘∘ ∘ ∘ ∘∘ ∘ ∘∘ ∘ ∘ ·
50 ┤ ·●────∘ ·∘· · ∘ ∘∘·∘∘ · ·∘∘∘∘∘∘∘∘∘∘ ·∘· ∘ ∘ ·∘· ∘ ·∘· · · ·∘· · · · · ·∘· · · · · · ·
│ ∘ ∘ ∘ ∘ ∘ ∘∘ ∘∘∘∘ ∘ ∘ ∘∘ ∘∘∘∘ ∘∘ ∘∘∘ ∘∘ · ∘ ·
│ ∘ ∘ ∘ ∘ ∘ ∘∘ ∘∘∘∘ ∘∘∘ ∘∘ ∘∘∘ ∘ ∘ ∘
│ ∘ ∘ ∘ ∘ ∘ ∘ ∘∘∘∘∘ ∘ ∘ ∘ · ·
25 ┤ ·∘· · ∘ · · ∘ ∘ · ·∘· · ∘∘∘ · ·∘∘∘· · · ∘∘· · · · ∘ · · · · · · · · · · · · · · · · ·
│ ∘ ∘ ∘ ∘ ∘ ∘· ∘∘∘ · ·
│ ∘ ∘ ∘ ∘ ∘ ∘
│ ∘ ∘ · ∘ · ·
│ ∘ ∘
0 ┤ ·∘· · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · ··· · · · · ·
│ ∘
│ · · ·
└─────────────────────┬─────────────────────────┬──────────────────────────┬───────────
10GB 100GB 1TB
# model score where RAM context get it 1 GPT-6 Astra 87±1 api $20.00/M 1.1M openai.com ↗ 2 Claude Opus 5.5 87±1 api $8.00/M 1M anthropic.com ↗ 3 GPT-6.1 Sol 85±1 api $4.00/M 1.1M openai.com ↗ 4 Claude Fable 5.1 85±1 api $20.00/M 1M anthropic.com ↗ 5 Claude Sonnet 5.5 84±1 api $4.00/M 1M claude.com ↗ 6 Claude Opus 5 83±1 api $10.00/M 1M anthropic.com ↗ 7 GPT-6 Sol 83±1 api $4.00/M 1.1M openai.com ↗ 8 GPT-5.5 Pro 82±1 api $67.50/M 1.1M openai.com ↗ 9 Gemini 4 Argon (limited p… 82±1 limited preview 10 GPT-5.6 Sol 82±1 api $8.00/M 1.1M openai.com ↗ 11 Claude Fable 5 81±1 api $20.00/M 1M anthropic.com ↗ 12 Claude Mythos Preview (li… 81±1† limited preview 1M 13 GPT-5.4 Pro 81±1 api $67.50/M 1.1M openai.com ↗ 14 GPT-5.5 80±1 api $11.25/M 1.1M openai.com ↗ 15 MiMo-V2.6-Pro 79±1 api $0.54/M +local ~443GB 1M mimo.mi.com ↗ 16 Hy4 preview (770B-A49B) 79±1 api $1.25/M +local ~248GB 1M tencent.com ↗ 17 GPT-5.6 Terra 79±1 api $4.50/M 1.1M openai.com ↗ 18 Qwen3.8 Max (2.4T-A95B) 78±1 api $3.00/M +local ~380GB 262k deepinfra.com ↗ 19 Claude Opus 4.8 78±0 api $10.00/M 1M anthropic.com ↗ 20 GPT-5.4 78±1 api $5.63/M 1.1M openai.com ↗ 21 DeepSeek V4.1 Flash 78±1 api $0.26/M +local ~537GB 1M deepseek.com ↗ 22 Claude Haiku 5.5 77±1 api $0.20/M 1M claude.com ↗ 23 Gemini 3.8 Flash 77±1 api $1.50/M 1M google.dev ↗ 24 Muse Spark 1.3 77±1 api $2.00/M 1M dev.meta.ai ↗ 25 Kimi K3 77±1 api $6.00/M +local ~884GB 1M moonshot.ai ↗ 26 GPT-5.2 Pro 77±1 api $57.75/M 400k openai.com ↗ 27 MiMo-V2.6-Flash (309B-A15… 77±1 api $0.17/M +local ~136GB 1M mimo.mi.com ↗ 28 GPT-6 Luna 77±1 api $0.20/M 1.1M openai.com ↗ 29 GPT-5.6 Luna 77±1 api $0.45/M 1.1M openai.com ↗ 30 Grok 4.6 77±1 api $3.00/M 500k x.ai ↗ 31 DeepSeek V4 Pro (0813) 76±1 api $1.63/M +local ~1010GB 1M deepinfra.com ↗ 32 Gemini 3.7 Flash 76±1 api $1.50/M 1M google.dev ↗ 33 GLM-5.3 76±1 api $2.15/M +local ~239GB 1M z.ai ↗ 34 Qwen3.8 Max (0902) 76±1 api $3.00/M 1M alibabacloud.co… 35 DeepSeek V3.2 Speciale 76±3† comparison only 36 DeepSeek V4 Flash Vision … 76±2 local ~104GB 1M llama.cpp › 37 Qwen3.8-Flash-Next 75±1 local ~59GB 262k llama.cpp › 38 Qwen3.8-Omni-Flash 75±2† api $0.23/M 1M alibabacloud.co… 39 Gemini 3.1 Pro (Preview) 75±1 api $4.50/M 1M google.dev ↗ 40 Muse Spark 1.2 75±1 api $2.00/M 1M dev.meta.ai ↗ 41 Claude Opus 4.7 75±1 api $10.00/M 1M anthropic.com ↗ 42 Grok 4.5 75±1 api $3.00/M 500k x.ai ↗ 43 Muse Spark 1.1 75±1 api $2.00/M 1M dev.meta.ai ↗ 44 Muse Spark 75±2 comparison only 45 Step 5 Preview 75±1 comparison only 46 Claude Sonnet 5 75±1 api $4.00/M 1M anthropic.com ↗ 47 Grok 4.7 75±1 api $3.00/M 500k x.ai ↗ 48 Claude Opus 4.6 75±1 api $10.00/M 1M anthropic.com ↗ 49 GPT-5.3-Codex 74±1 api $4.81/M 400k openai.com ↗ 50 DeepSeek V4 Flash (0731) 74±1 api $0.11/M +local ~92GB 1M deepinfra.com ↗ 51 Qwen3.7 Max 74±1 api $3.75/M 1M deepinfra.com ↗ 52 GLM-5.3-Flash 74±1 api $0.24/M +local ~103GB 1M z.ai ↗ 53 Gemini 3.5 Flash 74±1 api $3.38/M 1M google.dev ↗ 54 GPT-5.2 74±1 api $4.81/M 400k openai.com ↗ 55 Hunyuan Hy3 (295B-A21B) 74±1 api $0.23/M +local ~96GB 256k deepinfra.com ↗ 56 Gemini 3.6 Flash 74±1 api $1.50/M 1M google.dev ↗ 57 GLM-5.2 74±1 api $2.15/M +local ~121GB 1M z.ai ↗ 58 Dola Seed 2.0 Pro 74±2 comparison only 59 Qwen3.8 27B 73±1 api $1.60/M +local ~18GB 262k groq.com ↗ 60 Grok 4.20 Multi-Agent (03… 73±2 api $1.56/M 1M x.ai ↗ 61 Kimi K2.6 73±1 api $1.71/M +local ~317GB 262k moonshot.ai ↗ 62 GPT-5 Pro 73±1 api $41.25/M 400k openai.com ↗ 63 LongCat 2.0 (1.6T-A48B) 73±2 api $1.30/M +local ~2155GB 1M longcat.chat ↗ 64 Gemini 3 Pro (Preview) 73±1 comparison only 1M 65 GPT-5 72±1 api $3.44/M 400k openai.com ↗ 66 Qwen3.6 Max Preview 72±1 api $2.92/M 256k alibabacloud.co… 67 Claude Sonnet 4.6 72±1 api $6.00/M 1M anthropic.com ↗ 68 Gemini 3 Flash (Preview) 72±1 api $1.13/M 1M google.dev ↗ 69 DeepSeek V4 Pro (Preview,… 72±1 local ~604GB llama.cpp › 70 Claude Opus 4.5 72±1 api $10.00/M 200k anthropic.com ↗ 71 Qwen3.7 Plus 72±1 api $0.70/M 1M alibabacloud.co… 72 Kimi K2.7 Code 72±1 api $1.71/M +local ~359GB 262k kimi.ai ↗ 73 DeepSeek V4 Flash (Previe… 72±1 local ~103GB llama.cpp › 74 Grok 4.20 (0309 reasoning) 72±1 api $1.56/M 1M x.ai ↗ 75 Solar Open 2 (250B-A15B) 72±1† local ~527GB 1M how to run › 76 North Small Translate 1.0 71±6† api price unknown 16k cohere.com ↗ 77 GPT-5.4 mini 71±1 api $1.69/M 400k openai.com ↗ 78 Ling 3.0 Flash (124B-A5B) 71±2† api $0.090/M +loc… ~75GB 131k deepinfra.com ↗ 79 Mistral Large 4 (public p… 71±2 api $2.06/M 1M mistral.ai ↗ 80 GPT-5.1 71±1 api $3.44/M 400k openai.com ↗ 81 GLM-5.1 71±1 api $2.15/M +local ~119GB 200k z.ai ↗ 82 Qwen3.5 397B-A17B 71±0 api $1.09/M +local ~65GB 262k deepinfra.com ↗ 83 Ornith 1.5 35B-A3B 71±2† local ~22GB 262k LM Studio › 84 MiniMax-M3 70±1 api $0.48/M +local ~137GB 1M deepinfra.com ↗ 85 Grok 4.3 70±1 api $1.56/M 1M x.ai ↗ 86 Gemini 2.5 Pro Preview (0… 70±2 comparison only 87 o3-pro 70±2 api $35.00/M 200k openai.com ↗ 88 KAT-Coder-V2.5-Dev (35B-A… 70±2† local ~22GB 262k LM Studio › 89 Qwen3.6 Plus 70±1 api $1.13/M 1M together.ai ↗ 90 Inkling Small (266B-A12B) 70±1 api $0.64/M +local ~86GB 1M deepinfra.com ↗ 91 MiMo-V2-Pro 70±1 comparison only 92 Kimi K2.5 70±1 api $1.20/M +local ~160GB 262k aws.amazon.com ↗ 93 GLM-5 70±1 api $1.55/M +local ~119GB 205k z.ai ↗ 94 Qwen3-Max (2026-01-23) 70±1† comparison only 95 MiMo-V2.5-Pro (1T-A42B) 70±1 api $0.65/M +local ~163GB 1M novita.ai ↗ 96 Inkling (975B-A41B) 70±1 api $1.73/M +local ~156GB 1M deepinfra.com ↗ 97 Muse Glimmer 30B 70±1 api $0.64/M +local ~19GB 131k fireworks.ai ↗ 98 Nemotron 3 Ultra 550B-A55B 69±1 api $1.35/M +local ~89GB 1M together.ai ↗ 99 Qwen3.6 27B 69±1 api $1.20/M +local ~18GB 262k groq.com ↗ 100 Ring-2.6-1T 69±2† comparison only 101 A.X K2 (688B-A33B) 69±1† local ~1454GB 262k how to run › 102 o3 69±1 api $3.50/M 200k openai.com ↗ 103 MiMo-V2.5 (310B-A15B) 69±1 api $0.21/M +local ~100GB 1M novita.ai ↗ 104 Qwen3.5 Plus 69±2 api $0.90/M 1M alibabacloud.co… 105 Qwen3.5 122B-A10B 69±1 api $1.10/M +local ~75GB 262k alibabacloud.co… 106 Gemini 3.5 Flash-Lite 69±1 api $0.85/M 1M google.dev ↗ 107 GLM-4.7 69±1 api $1.00/M +local ~58GB 205k z.ai ↗ 108 GPT-5.5 Instant 69±1 comparison only 109 GPT-5.4 nano 69±1 api $0.46/M 400k openai.com ↗ 110 Claude Sonnet 4.5 69±1 api $6.00/M 200k anthropic.com ↗ 111 Qwen3.5 27B 69±0 api $0.84/M +local ~18GB 262k deepinfra.com ↗ 112 GLM-5V-Turbo 69±2 api $1.90/M 200k siliconflow.com… 113 GPT-5.2-Codex 69±2 comparison only 114 LongCat-Flash-Thinking (5… 69±1† comparison only 128k 115 Step 3.5 Flash (196B-A11B) 68±1 api $0.15/M +local ~118GB 256k stepfun.ai ↗ 116 DeepSeek V3.2 68±1 api $0.29/M +local ~110GB 128k deepinfra.com ↗ 117 MiMo-V2-Omni 68±2 comparison only 262k 118 Gemini 2.5 Pro Experiment… 68±3 comparison only 119 Grok 4 (0709) 68±1 comparison only 120 MiniMax-M2.7 68±1 api $0.53/M +local ~75GB 205k minimax.io ↗ 121 Qwen3.6 35B-A3B 68±1 api $0.31/M +local ~23GB 262k deepinfra.com ↗ 122 Ornith 1.0 35B (35B-A3B) 68±1 local ~23GB 262k LM Studio › 123 Hy3 preview (295B-A21B) 68±1 local ~629GB how to run › 124 GPT-5 mini 67±1 api $0.69/M 400k openai.com ↗ 125 Gemini 2.5 Pro 67±1 api $3.44/M 1M google.dev ↗ 126 Claude Opus 4 67±1 comparison only 127 K-EXAONE 2.0 (750B-A37B) 67±1† local ~791GB 262k how to run › 128 Claude Opus 4.1 67±1 comparison only 129 Kimi K2 Thinking 67±1 api $1.07/M +local ~160GB 262k aws.amazon.com ↗ 130 Laguna M.1 67±2 api price unknown… ~135GB 262k huggingface.co ↗ 131 Qwen3.7 Flash 67±1 api $0.055/M 1M alibabacloud.co… 132 Step 3.7 Flash (198B-A11B) 67±1 api $0.44/M +local ~104GB 256k stepfun.ai ↗ 133 Gemma 4 31B 67±1 api $0.19/M +local ~21GB 262k deepinfra.com ↗ 134 Gemini 3.1 Flash-Lite 67±1 api $0.56/M 1M google.dev ↗ 135 Qwen3-Coder-Next (80B-A3B) 67±2 api $0.60/M +local ~27GB 262k alibabacloud.co… 136 Gemini 2.5 Pro Preview (0… 67±2 comparison only 137 Gemini 2.5 Pro Preview (0… 67±1 comparison only 138 GigaChat 3.5 Ultra Reason… 67±1† local ~461GB 262k how to run › 139 Qwen3.5 35B-A3B 67±0 api $0.35/M +local ~23GB 262k deepinfra.com ↗ 140 DeepSeek V3.1 Terminus 67±1 local ~230GB 164k llama.cpp › 141 DeepSeek V3.2 Exp 67±1 local ~726GB how to run › 142 Qwen3 235B-A22B Thinking … 66±1 api $0.97/M +local ~76GB 131k novita.ai ↗ 143 Grok 4.1 Fast (Reasoning) 66±1 api $0.28/M 128k azure.microsoft… 144 o4-mini 66±1 api $1.93/M 200k openai.com ↗ 145 Qwen3.6 Flash 66±1 api $0.56/M 1M alibabacloud.co… 146 MiMo-V2-Flash (309B-A15B) 66±1 local ~108GB llama.cpp › 147 GLM-4.6 66±1 api $1.00/M +local ~58GB 205k z.ai ↗ 148 Kimi K2 Instruct (0711) 66±1 api $1.00/M +local ~160GB 131k novita.ai ↗ 149 o1 (2024-12-17) 66±1 api $26.25/M 200k openai.com ↗ 150 MiniMax-M2.5 66±1 api $0.53/M +local ~75GB 205k minimax.io ↗ 151 Qwen3-Max (2025-09-23) 66±1 api $2.40/M 262k alibabacloud.co… 152 DeepSeek V3.1 66±1 api $0.42/M +local ~110GB 131k deepinfra.com ↗ 153 Spark-X2.5 4B 66±1† local ~3.9GB 1M LM Studio › 154 MiniMax-M2.1 66±1 local ~79GB llama.cpp › 155 Nemotron Cascade 2 30B-A3B 66±1 local ~20GB 256k LM Studio › 156 Laguna XS 2.1 (33B-A3B) 66±2† local ~21GB 262k LM Studio › 157 LongCat-Flash-Chat (560B-… 66±1 local ~333GB 128k LM Studio › 158 o3-mini 65±1 api $1.93/M 200k openai.com ↗ 159 Claude 3.7 Sonnet 65±1 comparison only 160 Claude Sonnet 4 65±1 comparison only 161 MiMo-V2.6-Distill-Qwen-9B 65±2† local ~7.8GB LM Studio › 162 K-EXAONE (236B-A23B) 65±1 local ~152GB 262k llama.cpp › 163 DeepSeek R1 0528 65±1 api $0.91/M +local ~110GB 164k deepinfra.com ↗ 164 EXAONE 4.5 33B 65±2† local ~24GB 262k llama.cpp › 165 Qwen3.5 Flash 65±1 api $0.17/M 1M alibabacloud.co… 166 ChatGPT-4o (2025-03-26) 65±2 comparison only 167 INTELLECT-3 65±1 local ~48GB llama.cpp › 168 Nemotron 3 Super 120B-A12B 65±1 api $0.28/M +local ~73GB 262k aws.amazon.com ↗ 169 GLM-4.5 65±1 api $1.00/M +local ~58GB 131k z.ai ↗ 170 Solar Pro 4 65±2 api $0.53/M 524k upstage.ai ↗ 171 MiniMax-M2 65±1 api $0.53/M +local ~75GB 205k aws.amazon.com ↗ 172 Laguna XS.2 65±2 local ~21GB 262k LM Studio › 173 Qwen3 235B-A22B Instruct … 65±1 api $0.20/M +local ~76GB 262k deepinfra.com ↗ 174 Mistral Medium 3.5 (128B) 65±1 api $3.00/M +local ~42GB 262k mistral.ai ↗ 175 Command A+ (05-2026) 64±1 api price unknown… ~71GB 128k cohere.com ↗ 176 Claude Haiku 4.5 64±1 api $2.00/M 200k anthropic.com ↗ 177 Nemotron-Labs-3 Puzzle 75… 64±1† local ~35GB llama.cpp › 178 Qwen3-VL 235B-A22B Thinki… 64±1 local ~111GB 256k LM Studio › 179 Qwen3-Coder 480B-A35B Ins… 64±1 api $0.67/M +local ~78GB 262k novita.ai ↗ 180 Mistral Medium 3.1 (2508) 64±2 comparison only 181 Gemma 4 26B-A4B 64±1 api $0.14/M +local ~18GB 262k deepinfra.com ↗ 182 Kimi K2 Instruct (0905) 64±2 local ~298GB llama.cpp › 183 Qwen3-VL 235B-A22B Instru… 64±1 local ~134GB 256k LM Studio › 184 Nova 2 Pro (Preview) 64±2 api $3.78/M 1M aws.amazon.com ↗ 185 Grok 3 Mini Beta 64±1 comparison only 186 MiniCPM5 2B 64±1† local ~3GB 131k LM Studio › 187 Grok 3 Beta 63±1 comparison only 188 Qwen3-Next 80B-A3B Thinki… 63±1 api $0.41/M +local ~27GB 131k alibabacloud.co… 189 DeepSeek R1 (Jan '25) 63±1 api $4.00/M +local ~110GB 128k together.ai ↗ 190 Gemini 2.5 Flash Preview … 63±1 comparison only 191 Qwen3-Next 80B-A3B Instru… 63±1 api $0.34/M +local ~27GB 131k deepinfra.com ↗ 192 Granite 4.2 30B 63±1 api $0.28/M +local ~19GB 131k deepinfra.com ↗ 193 o1-preview 63±1 comparison only 194 Gemini 2.5 Flash Preview … 63±1 comparison only 195 GLM-4.5-Air 63±1 api $0.42/M +local ~35GB 131k z.ai ↗ 196 Qwen3.5 9B 63±1 api $0.11/M +local ~7.4GB 262k deepinfra.com ↗ 197 Seed-OSS 36B Instruct 63±1 local ~23GB 524k LM Studio › 198 GPT-4.5 Preview 63±1 comparison only 199 Trinity-Large-Thinking 63±2 api $0.39/M +local ~141GB 512k arcee.ai ↗ 200 Magistral Medium 1.2 (250… 63±2† comparison only 128k 201 MiniMax-M1 (80k) 62±1 api $0.96/M +local ~959GB 1M novita.ai ↗ 202 Gemma 4 12B 62±1 local ~8.5GB 262k LM Studio › 203 GPT-4.1 62±1 api $3.50/M 1M openai.com ↗ 204 Nemotron 3.5 Lightning 30… 62±2 api $0.087/M +loc… ~20GB 262k fireworks.ai ↗ 205 GLM-4.6V 62±2 api $0.45/M +local ~51GB 128k z.ai ↗ 206 Devstral 2 (2512) 62±2 local ~75GB 256k LM Studio › 207 Gemini 2.5 Flash Preview … 62±1 comparison only 208 A.X K1 (519B-A33B) 62±1† local ~1091GB how to run › 209 gpt-oss-120b 62±1 api $0.26/M +local ~67GB 131k groq.com ↗ 210 GPT-5 nano 62±1 api $0.14/M 400k openai.com ↗ 211 Qwen3-VL 32B Thinking 62±1 api $0.28/M +local ~22GB 256k alibabacloud.co… 212 Gemini 2.0 Flash-Lite (00… 62±2 comparison only 213 Mistral Large 3 (675B-A41… 62±1 api $0.75/M +local ~108GB 262k mistral.ai ↗ 214 Apriel 1.6 15B Thinker 62±2† local ~10GB LM Studio › 215 Gemini 2.5 Flash-Lite Pre… 62±2 comparison only 216 Gemini 2.5 Flash 62±1 api $0.85/M 1M google.dev ↗ 217 MiniMax-M1 (40k) 61±1† comparison only 218 Grok 4 Fast 61±2 comparison only 219 GLM-4.5V 61±2 api $0.90/M +local ~69GB 64k z.ai ↗ 220 Llama 3.1 Nemotron Ultra … 61±2 local ~82GB 128k llama.cpp › 221 HyperCLOVA X SEED 32B Thi… 61±2† comparison only 128k 222 Devstral Small 2 24B 61±3† local ~17GB 262k llama.cpp › 223 o1-mini 61±1 comparison only 224 Gemini 2.0 Pro Experiment… 61±1 comparison only 225 Qwen3 30B-A3B Thinking 25… 61±1 api $0.75/M +local ~20GB 262k alibabacloud.co… 226 Qwen3.5 4B 61±1 local ~4.3GB 262k llama.cpp › 227 Nemotron-Cascade-14B-Thin… 61±1† local ~10GB LM Studio › 228 Mellum2.1 12B-A2.5B Think… 60±2† local ~8.3GB LM Studio › 229 GPT-4.1 mini 60±1 api $0.70/M 1M openai.com ↗ 230 DeepSeek V3 0324 60±1 local ~108GB 164k llama.cpp › 231 Qwen3 30B-A3B Instruct 25… 60±1 api $0.15/M +local ~20GB 262k tokenfactory.ne… 232 Qwen3-VL 32B Instruct 60±1† api $0.28/M +local ~22GB 256k alibabacloud.co… 233 Falcon-H1R 7B 60±1† local ~6GB 262k LM Studio › 234 Gemini 2.0 Flash (001) 60±1 comparison only 235 ERNIE 4.5 300B-A47B 60±1† local ~97GB 131k llama.cpp › 236 Mistral Medium 3 (2505) 60±1 comparison only 237 Qwen3 235B-A22B (Apr 2025… 60±1 api $1.23/M +local ~76GB 131k alibabacloud.co… 238 Claude 3.5 Sonnet (2024-1… 60±1 comparison only 239 Granite 4.2 8B 60±1 api $0.11/M +local ~6.6GB 131k wandb.ai ↗ 240 Solar Open 100B (102B-A12… 60±1† local ~41GB 131k llama.cpp › 241 Qwen3-VL 30B-A3B Thinking 60±1† api $0.75/M +local ~21GB 256k alibabacloud.co… 242 Mistral Small 4 (119B-A6.… 60±2 api $0.26/M +local ~40GB 262k mistral.ai ↗ 243 Gemini 2.0 Flash-Lite Pre… 60±3 comparison only 244 QwQ-32B 59±1 local ~21GB 131k LM Studio › 245 OpenReasoning-Nemotron-32B 59±2† local ~21GB LM Studio › 246 DeepSeek V3 (Dec '24) 59±1 api $0.46/M +local ~337GB 131k deepinfra.com ↗ 247 Gemma 2 9B IT SimPO 59±1 local ~7.6GB llama.cpp › 248 Gemma 2 9B IT DPO (Prince… 59±1 local ~7.6GB llama.cpp › 249 Trinity-Large-Preview 59±2 local ~155GB llama.cpp › 250 GLM-4.7-Flash 59±1 api $0.0000/M +lo… ~19GB 200k z.ai ↗ 251 EXAONE 4.0 32B 59±1 local ~20GB 131k LM Studio › 252 Phi-4-reasoning-plus 14B 59±1 local ~10GB LM Studio › 253 Nemotron-Cascade-8B 59±1† local ~6.3GB LM Studio › 254 Qwen3 14B 59±1 api $0.61/M +local ~10GB 33k alibabacloud.co… 255 Mercury 2 59±2 comparison only 256 DiffusionGemma 26B-A4B 59±1† comparison only 262k 257 gpt-oss-20b 58±1 api $0.13/M +local ~14GB 131k groq.com ↗ 258 Llama 3.3 Nemotron Super … 58±2 local ~22GB llama.cpp › 259 Olmo 3.1 32B Think 58±1 local ~21GB 66k LM Studio › 260 DeepSeek R1 Distill Llama… 58±1 api $0.80/M +local ~30GB 131k novita.ai ↗ 261 Phi-4-reasoning 14B 58±1 api $0.22/M +local ~10GB 33k azure.microsoft… 262 Olmo 3 32B Think 58±1 local ~21GB 66k LM Studio › 263 Magistral Small 1.2 58±1 api $0.75/M +local ~16GB 131k aws.amazon.com ↗ 264 Nemotron-Cascade-8B-Think… 58±2† local ~6.3GB LM Studio › 265 Qwen3-VL 30B-A3B Instruct 58±1† api $0.35/M +local ~21GB 256k alibabacloud.co… 266 Qwen3-VL 8B Thinking 58±1 api $0.66/M +local ~6.8GB 256k alibabacloud.co… 267 Qwen3 32B 58±0 api $0.13/M +local ~20GB 131k deepinfra.com ↗ 268 Qwen3 Coder 30B-A3B Instr… 58±1 api $0.12/M +local ~20GB 160k novita.ai ↗ 269 Nova 2 Lite 58±2 api $0.94/M 1M aws.amazon.com ↗ 270 Llama 3.3 Nemotron Super … 58±1 local ~18GB 131k llama.cpp › 271 Nemotron 3 Nano 30B-A3B 58±1 api $0.087/M +loc… ~20GB 262k deepinfra.com ↗ 272 Ministral 3 14B Reasoning… 58±1 local ~10GB 262k llama.cpp › 273 GPT-4o (2024-08-06) 57±1 api $4.38/M 128k openai.com ↗ 274 Gemini 2.0 Flash Thinking… 57±2 comparison only 275 DeepSeek R1 0528 Qwen3 8B 57±1 api $0.068/M +loc… ~6.3GB 128k novita.ai ↗ 276 Granite 4.2 3B 57±1 api $0.052/M +loc… ~3.7GB 131k deepinfra.com ↗ 277 Command A (03-2025) 57±1 api $4.38/M +local ~37GB 256k azure.microsoft… 278 GPT-4o (2024-11-20) 57±1 api $4.38/M 128k openai.com ↗ 279 Gemini 2.5 Flash-Lite 57±2 api $0.17/M 1M google.dev ↗ 280 Qwen2.5-Max (2025-01-25) 57±1 comparison only 281 Qwen2.5-Omni 7B 57±2 local ~6.4GB llama.cpp › 282 GPT-4o (2024-05-13) 57±0 api $7.50/M 128k openai.com ↗ 283 QED-Nano 57±2 local ~4.1GB 262k llama.cpp › 284 Llama 4 Maverick (402B-A1… 57±1 api $0.42/M +local ~65GB 1M aws.amazon.com ↗ 285 Claude 3.5 Sonnet (2024-0… 57±1 comparison only 286 Qwen2.5 Coder 32B Instruct 57±1 api $0.74/M +local ~21GB 131k cloudflare.com ↗ 287 Qwen3 4B Thinking 2507 57±1 local ~3.9GB 262k LM Studio › 288 Gemini 1.5 Pro Experiment… 56±1 comparison only 289 Gemini 1.5 Pro (002) 56±1 comparison only 290 Athene-V2-Chat 72B 56±2 local ~44GB 33k LM Studio › 291 Gemini 2.0 Flash Experime… 56±1 comparison only 292 Sky-T1 32B Preview 56±5 local ~21GB LM Studio › 293 Qwen Plus (2025-01-25) 56±1 api $0.60/M 131k alibabacloud.co… 294 Grok 2 (1212) 56±1 comparison only 295 Qwen3-VL 4B Thinking 56±1 local ~4.1GB 256k llama.cpp › 296 EXAONE Deep 32B 56±2† local ~22GB 33k Ollama › 297 DeepSeek R1 Distill Qwen … 56±1 api $1.59/M +local ~21GB 131k cloudflare.com ↗ 298 Qwen2.5-VL 72B Instruct 56±1 local ~46GB 33k LM Studio › 299 Granite 4.1 30B 56±1† local ~19GB Ollama › 300 Mistral Small 3.2 24B 56±1 api $0.11/M +local ~16GB 128k deepinfra.com ↗ 301 DeepSeek-V2.5 (1210) 56±1 local ~82GB llama.cpp › 302 Qwen3 8B 56±1 api $0.31/M +local ~6.1GB 33k alibabacloud.co… 303 DeepSeek R1 Distill Qwen … 56±1 local ~10GB 131k LM Studio › 304 Ministral 3 14B 56±1 api $0.20/M +local ~9.1GB 262k mistral.ai ↗ 305 Nemotron Nano 9B v2 56±1 api $0.10/M +local ~7.6GB 131k aws.amazon.com ↗ 306 Magistral Medium 1.0 (250… 56±1 comparison only 307 GPT-4 Turbo (2024-04-09) 55±0 api $15.00/M 128k openai.com ↗ 308 Yi-Lightning 55±1 api price unknown 16k lingyiwanwu.com… 309 Qwen2.5 72B Instruct 55±1 api $0.37/M +local ~30GB 33k deepinfra.com ↗ 310 Mistral Large 2 (2411) 55±1 local ~78GB 128k llama.cpp › 311 Ministral 3 8B Reasoning … 55±1 local ~6.5GB 262k LM Studio › 312 Magistral Small 1.0 (2506) 55±1 local ~16GB 128k LM Studio › 313 Qwen3 30B-A3B (Apr 2025, … 55±1 api $0.35/M +local ~19GB 33k alibabacloud.co… 314 Molmo 2 8B 55±2 local ~6.8GB llama.cpp › 315 Magistral Medium 1.1 (250… 55±2† comparison only 40k 316 Gemini 1.5 Pro Experiment… 55±1 comparison only 317 Qwen3 4B Instruct 2507 55±1 local ~3.9GB 262k LM Studio › 318 Qwen2.5 Coder 14B Instruct 55±1 local ~10GB 131k LM Studio › 319 Qwen3-VL 8B Instruct 55±1 api $0.31/M +local ~6.8GB 256k alibabacloud.co… 320 Spark-X2.5 1.7B 55±1† local ~2.7GB 1M llama.cpp › 321 GigaChat 3.5 Ultra Instru… 55±1† local ~457GB 262k how to run › 322 Llama 3.1 405B Instruct 55±1 api $2.40/M +local ~160GB 128k aws.amazon.com ↗ 323 Qwen2.5 32B Instruct 55±1 local ~21GB 131k LM Studio › 324 Command A Reasoning (08-2… 55±2 local ~34GB llama.cpp › 325 Bielik 11B v3.0 Instruct 55±2 local ~8.1GB LM Studio › 326 Olmo 3.1 32B Instruct 55±2 local ~21GB 66k LM Studio › 327 OpenThinker 32B 55±1† local ~21GB 33k LM Studio › 328 Magistral Small 1.1 (2507) 55±2† local ~16GB 40k LM Studio › 329 EXAONE Deep 7.8B 55±2† local ~6.5GB 33k llama.cpp › 330 Qwen2.5-Math 72B Instruct 55±2† local ~44GB 4k LM Studio › 331 OpenThinker2 32B 54±1 local ~21GB 33k LM Studio › 332 Falcon-H1 34B Instruct 54±3 local ~23GB 262k llama.cpp › 333 Llama 3.3 70B Instruct 54±1 api $0.72/M +local ~24GB 128k aws.amazon.com ↗ 334 Llama 4 Scout (109B-A17B) 54±1 api $0.15/M +local ~36GB 328k deepinfra.com ↗ 335 Nemotron 3 Nano 4B 54±1 local ~3.8GB 262k LM Studio › 336 Granite 4.1 8B 54±1 api $0.11/M +local ~7GB 131k replicate.com ↗ 337 Pixtral Large (2411) 54±3 local ~51GB llama.cpp › 338 Bielik 11B v2.6 Instruct 54±3 local ~8.1GB LM Studio › 339 Nova Premier 54±2† api $5.00/M 1M aws.amazon.com ↗ 340 Mellum2 12B-A2.5B Thinking 54±1† local ~8.6GB 131k LM Studio › 341 Mistral Large 2 (2407) 54±1 local ~45GB 128k llama.cpp › 342 Mistral Small 3.1 24B (25… 54±1 local ~17GB 128k llama.cpp › 343 GPT-4 Turbo Preview (1106) 54±1 comparison only 344 GPT-4 Turbo Preview (0125) 54±1 comparison only 345 Claude 3.5 Haiku 54±1 comparison only 346 Llama 3.1 Tulu 3 70B 54±1 local ~43GB 131k LM Studio › 347 DeepSeek-V2.5 54±1 local ~151GB 164k Ollama › 348 DeepSeek-V2 Chat (0628) 54±1 local ~66GB llama.cpp › 349 GLM-4 9B Chat 1M 53±5 local ~8.1GB 1M llama.cpp › 350 Mistral Small 3 24B (2501) 53±1 local ~16GB 32k LM Studio › 351 Qwen2-VL 72B Instruct 53±1 local ~34GB llama.cpp › 352 Gemini 1.5 Flash Experime… 53±1 comparison only 353 GPT-4o mini (2024-07-18) 53±0 api $0.26/M 128k openai.com ↗ 354 Gemini 1.5 Pro (001) 53±1 comparison only 355 Llama 3.1 Tulu 3 70B DPO 53±1 local ~22GB 131k llama.cpp › 356 Qwen Turbo (2024-11-01) 53±2 comparison only 357 Qwen2.5-VL 32B Instruct 53±2 local ~22GB 33k LM Studio › 358 DeepSeek-Coder-V2 Instruc… 53±1 local ~92GB llama.cpp › 359 s1.1 32B 53±1 local ~22GB 33k LM Studio › 360 DeepSeek-Coder-V2 Instruc… 53±1 local ~151GB llama.cpp › 361 Olmo 3 7B Think 53±1 local ~5.8GB 66k LM Studio › 362 GPT-4.1 nano 53±1 api $0.17/M 1M openai.com ↗ 363 Gemini 1.5 Flash (002) 53±1 comparison only 364 Gemma 3 27B 53±1 api $0.10/M +local ~18GB 131k deepinfra.com ↗ 365 Phi-4 14B 53±1 api $0.087/M +loc… ~10GB 16k deepinfra.com ↗ 366 Gemma 4 E4B 53±1 local ~6.9GB 131k LM Studio › 367 Ministral 3 8B 53±1 api $0.15/M +local ~6.5GB 262k mistral.ai ↗ 368 Qwen3-VL 4B Instruct 53±1 local ~4.1GB 256k llama.cpp › 369 Claude 3 Opus 53±0 comparison only 370 LFM2.5 2.6B 53±1 local ~3.2GB 131k LM Studio › 371 Bielik 11B v2.1 Instruct 53±3 local ~8.5GB llama.cpp › 372 Llama 3.1 70B Instruct 52±1 api $0.72/M +local ~27GB 128k aws.amazon.com ↗ 373 Yi-Large Preview 52±1 comparison only 374 Qwen2.5-Math 7B Instruct 52±2† local ~6GB 4k LM Studio › 375 LFM2.5 8B-A1B 52±1 local ~6.5GB 128k LM Studio › 376 GPT-4 (0314) 52±1 comparison only 377 GLM-4 (0520) 52±1 comparison only 378 ERNIE 4.5 21B-A3B 52±1 comparison only 379 Bielik 11B v2.2 Instruct 52±3 local ~8.5GB llama.cpp › 380 Granite 4.0 H Small (32B-… 52±1 api $0.11/M +local ~21GB 131k replicate.com ↗ 381 Llama 3.1 Nemotron 70B In… 52±1 local ~43GB 128k LM Studio › 382 Palmyra X4 (palmyra-x-004) 52±4 comparison only 383 Nemotron-4 340B Instruct 52±2 comparison only 384 Aya Expanse 32B 52±2 api $0.75/M +local ~22GB 128k cohere.com ↗ 385 Gemma 3 12B 52±1 api $0.075/M +loc… ~9.1GB 131k deepinfra.com ↗ 386 Qwen2.5 Coder 7B Instruct 52±1 local ~6GB 131k LM Studio › 387 Qwen2.5 14B Instruct 52±1 local ~10GB 131k LM Studio › 388 Cogito v1 Preview Llama 8B 51±5 local ~6.2GB 128k LM Studio › 389 Hermes 3 Llama 3.1 405B 51±1 local ~114GB 131k llama.cpp › 390 GPT-4 (0613) 51±1 api $37.50/M 8k openai.com ↗ 391 OpenCoder-8B-Instruct 51±1 local ~6.4GB llama.cpp › 392 Command A Translate (08-2… 51±5† api price unknown 8k cohere.com ↗ 393 Llama 3.1 Nemotron 51B In… 51±2 local ~34GB llama.cpp › 394 Ministral 3 3B Reasoning … 51±1 local ~3.5GB 262k LM Studio › 395 Qwen3 4B 51±1 local ~3.7GB 33k LM Studio › 396 Yi-Large 51±1 comparison only 397 Llama-3-Instruct-8B-SimPO… 51±1 local ~6.6GB llama.cpp › 398 LFM2 24B-A2B 51±1 local ~16GB 33k LM Studio › 399 AceReason-Nemotron-1.0-14B 51±2† local ~10GB LM Studio › 400 Llama-3-Instruct-8B-SimPO 51±1 local ~6.6GB llama.cpp › 401 Athene-70B 51±1 local ~29GB llama.cpp › 402 Qwen2 72B Instruct 51±1 local ~33GB 131k llama.cpp › 403 Bielik 11B v2.0 Instruct 51±3 local ~8.5GB llama.cpp › 404 OpenThinker3 7B 50±1 local ~6GB 33k LM Studio › 405 CodeQwen1.5 7B Chat 50±1 local ~6.4GB 64k llama.cpp › 406 Jamba 1.5 Large 50±1 local ~839GB how to run › 407 Gemini 1.5 Flash (001) 50±1 comparison only 408 Llama-3-8B-ProLong-512k-I… 50±5 local ~35GB how to run › 409 EXAONE Deep 2.4B 50±2† local ~3.2GB 33k Ollama › 410 Mellum2 12B-A2.5B Instruct 50±2† local ~8.6GB 131k LM Studio › 411 OpenCodeInterpreter-DS-33B 50±1 local ~22GB LM Studio › 412 Llama 3.2 90B Vision Inst… 50±1 local ~58GB 128k Ollama › 413 DeepSeek-Coder 33B Instru… 50±1 local ~21GB 16k LM Studio › 414 Yi 34B 200K (base) 50±5 local ~23GB 200k llama.cpp › 415 Gemma 2 27B IT 50±1 local ~18GB 8k LM Studio › 416 Bielik 11B v2.3 Instruct 50±2 local ~8.6GB llama.cpp › 417 Qwen2.5-VL 7B Instruct 50±1 local ~6.4GB llama.cpp › 418 Molmo 72B (0924) 50±2 local ~309GB how to run › 419 Codestral 22B v0.1 50±1 local ~15GB 32k LM Studio › 420 GLM-4 (0116) 50±1 comparison only 421 Llama 3 70B Instruct 50±0 local ~24GB 8k llama.cpp › 422 WizardLM-2 8x22B 50±1 local ~91GB Ollama › 423 Nova Pro 50±1 api $1.40/M 300k aws.amazon.com ↗ 424 WizardCoder 33B V1.1 50±1 local ~22GB 16k llama.cpp › 425 Mistral Small Instruct 22… 49±2 local ~15GB 32k LM Studio › 426 Qwen2-Math 72B Instruct 49±1 local ~32GB 4k llama.cpp › 427 OLMo 2 32B Instruct (0325) 49±1 local ~21GB LM Studio › 428 OpenThinker2 7B 49±2† local ~6.4GB 33k llama.cpp › 429 Aya 23 35B 49±3 local ~22GB 8k LM Studio › 430 Granite 4.1 3B 49±1† local ~3.7GB Ollama › 431 OpenCodeInterpreter-DS-6.… 49±1 local ~5.8GB llama.cpp › 432 Artigenz-Coder-DS-6.7B 49±1 local ~5.8GB llama.cpp › 433 LFM2.5 1.2B Thinking 49±2 local ~2.2GB 33k LM Studio › 434 Qwen3-VL 2B Thinking 49±1 local ~2.7GB 256k llama.cpp › 435 speechless-codellama-34b-… 49±1 local ~23GB llama.cpp › 436 Pixtral 12B (2409) 49±2 local ~9.6GB llama.cpp › 437 Nova Lite 49±2 api $0.10/M 300k aws.amazon.com ↗ 438 MegaBeam-Mistral-7B-512k 49±4 local ~6.1GB llama.cpp › 439 GPT-4o Audio Preview (202… 49±2 comparison only 440 Claude 3 Sonnet 49±1 comparison only 441 AceReason-Nemotron-1.1-7B 49±2† local ~6GB LM Studio › 442 Mixtral 8x22B Instruct v0… 49±1 local ~56GB 64k llama.cpp › 443 Yi-Coder 9B Chat 49±1 local ~6.8GB 128k LM Studio › 444 Qwen2-Math 7B Instruct 49±2† local ~6GB 4k LM Studio › 445 Magicoder-S-DS 6.7B 49±1 local ~5.8GB llama.cpp › 446 InternLM2.5 7B Chat 49±2 local ~6.1GB LM Studio › 447 Hermes 3 Llama 3.1 70B 49±1 local ~44GB 131k Ollama › 448 Qwen2.5 7B Instruct 48±1 api $0.30/M +local ~6GB 131k together.ai ↗ 449 DeepSeek R1 Distill Qwen … 48±1 local ~6GB 131k LM Studio › 450 XwinCoder-34B 48±1 local ~23GB llama.cpp › 451 StarChat2 15B v0.1 48±1 local ~12GB 16k llama.cpp › 452 Code Millenials 34B 48±1 local ~23GB llama.cpp › 453 Reka Core (2024-05-01) 48±1 comparison only 454 Gemini 1.5 Flash-8B (001) 48±2 comparison only 455 T5 Large 48±9 local ~4.6GB 1k how to run › 456 WaveCoder-Ultra-6.7B 48±1 local ~5.8GB llama.cpp › 457 Phi-3.5-MoE-instruct 48±1 local ~26GB 128k LM Studio › 458 Claude 3 Haiku 48±0 comparison only 459 GPT-4o mini Audio Preview… 48±2 comparison only 460 Gemma 3n E4B 48±1 local ~6.2GB 32k llama.cpp › 461 Hermes 2 Theta Llama-3 70B 48±2 local ~47GB 8k llama.cpp › 462 Granite 4.0 H Tiny (7B-A1… 48±1 local ~5.6GB 131k LM Studio › 463 DeepSeek-Coder 7B Instruc… 48±1 local ~5.9GB llama.cpp › 464 Qwen2 57B-A14B Instruct 48±2 local ~35GB 33k LM Studio › 465 Qwen2.5 Coder 3B Instruct 48±1 local ~3.3GB 33k LM Studio › 466 DeepSeek-Coder 6.7B Instr… 48±1 local ~5.8GB 16k llama.cpp › 467 MiniCPM3-4B 48±2 local ~4GB 32k LM Studio › 468 Bielik 4.5B v3.0 Instruct 48±2 local ~4.3GB LM Studio › 469 Qwen3.5 2B 47±1 local ~2.9GB 262k llama.cpp › 470 Gemma 2 9B IT 47±1 local ~7GB 8k LM Studio › 471 Command R Plus (08-2024) 47±1 api $4.38/M +local ~63GB 128k cohere.com ↗ 472 GLM-4 9B Chat 47±2 local ~7.9GB 128k Ollama › 473 Yi-1.5 34B (base) 47±3 local ~22GB 4k LM Studio › 474 Phi-4-mini-reasoning 3.8B 47±2 api $0.13/M +local ~3.8GB 128k azure.microsoft… 475 Qwen1.5 110B Chat 47±1 local ~45GB 32k llama.cpp › 476 WhiteRabbitNeo-33B-v1 47±1 local ~22GB llama.cpp › 477 DeepSeek R1 Distill Llama… 47±1 local ~6.2GB 131k LM Studio › 478 EXAONE 4.0 1.2B 47±2† local ~2.2GB 66k LM Studio › 479 InternVL2-26B 47±2 local ~55GB how to run › 480 speechless-coder-ds-6.7b 47±1 local ~5.8GB llama.cpp › 481 Gemma 4 E2B 47±1 api $0.050/M +loc… ~5.1GB 131k aws.amazon.com ↗ 482 Magicoder-S-CL-7B 47±1 local ~5.8GB llama.cpp › 483 Aya Expanse 8B 47±2 local ~6.2GB 8k LM Studio › 484 Phind CodeLlama 34B v2 47±1 local ~23GB 16k Ollama › 485 CogVLM2 Llama3 Chinese Ch… 47±7† comparison only 486 InternLM2.5 7B Chat 1M 47±4 local ~6.4GB llama.cpp › 487 EuroLLM 9B Instruct 47±3 local ~6.9GB 4k LM Studio › 488 Granite 4.0 H Micro 3B 47±1 local ~3.4GB 131k LM Studio › 489 GPT-3.5 Turbo (0125) 46±1 api $0.75/M 16k openai.com ↗ 490 Mistral Large (2402) 46±1 comparison only 491 Jamba 1.5 Mini 46±1 local ~34GB llama.cpp › 492 MedGemma 4B IT 46±3 local ~4.1GB llama.cpp › 493 OLMo 2 13B Instruct (1124) 46±1 local ~9.6GB 4k LM Studio › 494 Falcon-H1 7B Instruct 46±3 local ~6GB 262k LM Studio › 495 GPT-3.5 Turbo (1106) 46±1 comparison only 496 Cogito v1 Preview Llama 3B 46±4 local ~3.4GB 128k LM Studio › 497 Code Llama 70B Instruct 46±1 local ~28GB 16k llama.cpp › 498 Hermes 2 Pro Llama 3 8B 46±3 local ~6.6GB 8k llama.cpp › 499 Llama 3 70B (base) 46±2 local ~43GB LM Studio › 500 Granite 4.0 Micro 3B 46±1 local ~3.5GB 131k LM Studio › 501 Command R (08-2024) 46±1 api $0.26/M +local ~21GB 128k cohere.com ↗ 502 GPT-3.5 Turbo (0301) 46±1 comparison only 503 Nova Micro 46±2 api $0.061/M 128k aws.amazon.com ↗ 504 T5 Base 46±9 local ~2.4GB 1k how to run › 505 OpenChat 3.5 0106 46±1 local ~6.1GB 8k llama.cpp › 506 Qwen2-VL 7B Instruct 46±1 local ~6.4GB LM Studio › 507 Qwen2.5-Math 1.5B Instruct 46±2† local ~2.4GB 4k LM Studio › 508 Yi 1.5 34B Chat 46±1 local ~22GB 4k LM Studio › 509 DBRX Instruct 46±1 local ~85GB 32k Ollama › 510 Granite 34B Code Instruct… 46±1 local ~22GB Ollama › 511 WizardCoder Python 34B V1… 46±1 local ~23GB 16k Ollama › 512 Gemini 1.0 Pro (001) 46±2 comparison only 513 Mistral Medium (2312) 46±1 comparison only 514 GPT-3.5 Turbo (0613) 46±1 comparison only 515 Granite 3.3 8B Instruct 46±1 local ~6.3GB 128k LM Studio › 516 Tulu 2 70B 46±2† local ~28GB llama.cpp › 517 LFM2 8B-A1B 46±1 comparison only 518 OpenChat 3.5 0106 Gemma 46±3 local ~6.9GB 8k llama.cpp › 519 Olmo 3 7B Instruct 46±1 local ~5.8GB 66k LM Studio › 520 InternLM2.5 20B Chat 46±2 local ~14GB 33k llama.cpp › 521 StarCoder2-15B-Instruct-v… 46±1 local ~11GB 16k Ollama › 522 Qwen1.5 72B Chat 45±1 local ~31GB 32k llama.cpp › 523 speechless-starcoder2-15b 45±1 local ~12GB llama.cpp › 524 Ministral 3 3B 45±1 api $0.10/M +local ~3.5GB 262k mistral.ai ↗ 525 Claude 2.0 45±1 comparison only 526 OpenThinker 7B 45±2† local ~6GB 33k LM Studio › 527 Phi-3 Medium 128K Instruct 45±1 local ~9.8GB 128k LM Studio › 528 QwQ-32B-Preview 45±2 local ~21GB 33k LM Studio › 529 DeepSeek-Coder-V2 Lite In… 45±1 local ~11GB 128k LM Studio › 530 GPT-4o Audio Preview (202… 45±3 comparison only 531 Claude 2.1 45±1 comparison only 532 Zephyr ORPO 141B-A35B v0.1 45±2 local ~56GB 66k llama.cpp › 533 Phi-3-medium-4k-instruct 45±1 local ~9.8GB 4k LM Studio › 534 Qwen2-Math 1.5B Instruct 45±2† local ~2.4GB 4k LM Studio › 535 Mistral Small (2402) 45±1 comparison only 536 Llama 3.1 8B Instruct 45±0 api $0.22/M +local ~6.2GB 128k aws.amazon.com ↗ 537 Command R Plus (04-2024) 45±1 local ~43GB 128k llama.cpp › 538 Mistral NeMo Instruct (24… 44±1 local ~8.7GB 128k LM Studio › 539 LLaVA-OneVision Qwen2 72B… 44±3 local ~155GB how to run › 540 Phi-3-mini-4k-instruct (J… 44±1 local ~3.7GB LM Studio › 541 Yi 1.5 9B Chat 44±1 local ~6.8GB 4k LM Studio › 542 Command R7B (12-2024) 44±1 local ~6.2GB 128k LM Studio › 543 Qwen2-Audio 7B Instruct 44±3 local ~6.5GB llama.cpp › 544 Gemma 3 4B 44±1 api $0.063/M +loc… ~3.8GB 131k deepinfra.com ↗ 545 Eurus-2-7B-PRIME 44±2 local ~6.4GB 4k llama.cpp › 546 Llama 3.1 Tulu 3 8B 44±1 local ~6.2GB 131k LM Studio › 547 Qwen2.5-VL 3B Instruct 44±1 local ~3.5GB llama.cpp › 548 Qwen1.5 32B Chat 44±1 local ~22GB 32k llama.cpp › 549 Granite 20B Code Instruct… 44±1 local ~14GB Ollama › 550 Qwen2 7B Instruct 44±1 local ~6GB 131k LM Studio › 551 Falcon3 7B Instruct 44±3 local ~5.9GB 32k LM Studio › 552 Xwin-LM 70B v0.1 44±1 local ~28GB llama.cpp › 553 Qwen3-VL 2B Instruct 44±1 local ~2.7GB 256k llama.cpp › 554 Dolphin 2.6 Mixtral 8x7B 44±1 local ~29GB llama.cpp › 555 Granite 8B Code Instruct … 44±1 local ~6.6GB llama.cpp › 556 Falcon3 10B Instruct 44±3 local ~7.6GB 32k LM Studio › 557 Code Tulu 2 34B 44±2† local ~72GB how to run › 558 Qwen3 1.7B 44±1 local ~2.6GB 33k LM Studio › 559 Phi-4-mini 3.8B 44±1 api $0.13/M +local ~3.8GB 128k azure.microsoft… 560 Ministral 8B (2410) 44±1 local ~6.2GB 128k LM Studio › 561 Granite 3.2 8B Instruct 44±1 local ~6.6GB 131k llama.cpp › 562 Code Llama 70B Python 44±1 local ~28GB 16k llama.cpp › 563 Llama 3.1 8B (base) 44±3 local ~6.2GB LM Studio › 564 LFM2.5 1.2B Instruct 44±2 local ~2.2GB 33k LM Studio › 565 Granite 3.0 8B Instruct 43±1 local ~6.6GB 4k llama.cpp › 566 Phi-3.5-mini-instruct 43±1 local ~3.7GB 128k LM Studio › 567 Hermes 2 Theta Llama 3 8B 43±2 local ~6.6GB 8k llama.cpp › 568 Yi 34B Chat 43±1 local ~23GB 4k llama.cpp › 569 CogVLM2 Llama3 Chat 19B 43±3 local ~43GB how to run › 570 Code Llama 70B (base) 43±1 local ~28GB 16k llama.cpp › 571 Gemma 3n E2B IT 43±2 local ~4.7GB 32k llama.cpp › 572 Reka Flash 21B (2024-02-2… 43±1 comparison only 573 T5 Small 43±9 local ~1.8GB 1k how to run › 574 MiniCPM-V 2.6 43±2 local ~6.4GB llama.cpp › 575 Qwen2.5 Coder 1.5B Instru… 43±1 local ~2.4GB 33k LM Studio › 576 LFM2 2.6B 43±2 local ~3GB 33k LM Studio › 577 SmolLM3 3B 43±1 local ~3.3GB 66k LM Studio › 578 Granite 3.1 8B Instruct 43±1 local ~6.6GB 128k llama.cpp › 579 Llama 3.2 11B Vision Inst… 43±2 local ~8.8GB 128k Ollama › 580 Qwen2.5 3B Instruct 43±1 local ~3.3GB 33k LM Studio › 581 DeepSeek-Coder 1.3B Instr… 43±1 local ~2.4GB 16k Ollama › 582 Aya 23 8B 43±3 local ~6.2GB 8k LM Studio › 583 Phi-3-small-8k-instruct 43±2 local ~17GB how to run › 584 CodeGemma 7B IT (v1.0) 43±1 local ~7.1GB llama.cpp › 585 Aya Vision 32B 42±3 api price unknown… ~71GB 16k cohere.com ↗ 586 Tulu 2 DPO 70B 42±1 local ~32GB llama.cpp › 587 Mathstral 7B v0.1 42±1 local ~5.8GB 32k LM Studio › 588 Yi-1.5 9B (base) 42±2 local ~6.8GB 4k LM Studio › 589 Claude Instant 1.2 42±1 comparison only 590 Llama 3 8B Instruct 42±0 local ~6.6GB 8k llama.cpp › 591 WizardLM 70B V1.0 42±1 local ~32GB llama.cpp › 592 WizardCoder Python 13B V1… 42±2 local ~9.8GB 16k llama.cpp › 593 Hermes 3 Llama 3.1 8B 42±1 local ~6.6GB 128k llama.cpp › 594 Falcon2 11B (base) 42±3 local ~8.2GB 8k Ollama › 595 Granite 34B Code Base (8K) 42±1 local ~24GB llama.cpp › 596 speechless-starcoder2-7b 42±1 local ~6.1GB llama.cpp › 597 Command R (03-2024) 42±1 local ~24GB 128k llama.cpp › 598 Ministral 3B (2410) 42±1 comparison only 599 OLMo 2 7B Instruct (1124) 42±1 local ~5.8GB 4k LM Studio › 600 WizardCoder-15B-V1.0 42±1 local ~12GB llama.cpp › 601 Snowflake Arctic Instruct 41±2 local ~1007GB how to run › 602 Phi-3-vision-128k-instruct 41±2 local ~10GB how to run › 603 OpenChat 3.5 41±2 local ~6.1GB 8k llama.cpp › 604 InternVL2-4B 41±2 local ~10GB how to run › 605 Qwen1.5 14B Chat 41±1 local ~11GB 32k Ollama › 606 Code Llama 34B Python 41±1 local ~23GB 16k Ollama › 607 speechless-coding-7b-16k-… 41±1 local ~16GB how to run › 608 DeepSeek-Coder 33B Base 41±1 local ~22GB 16k llama.cpp › 609 InternLM2 Chat 20B 41±1 local ~14GB llama.cpp › 610 DeepSeek LLM 67B Chat 41±1 local ~31GB 4k llama.cpp › 611 Llama 2 70B (base) 41±1 local ~32GB 4k llama.cpp › 612 Mixtral 8x7B v0.1 (base) 41±1 local ~29GB 32k llama.cpp › 613 North Micro Vision Instru… 41±2† local ~3.8GB 128k how to run › 614 CodeQwen1.5 7B (base) 41±1 local ~6.4GB 66k llama.cpp › 615 Granite 4.0 1B 41±1 local ~2.4GB 128k LM Studio › 616 Mixtral 8x7B Instruct v0.1 41±1 local ~29GB 32k LM Studio › 617 Molmo 7B-D (0924) 41±1 local ~35GB how to run › 618 OpenHermes-2.5-Code-290k-… 41±1 local ~9.8GB llama.cpp › 619 Granite 4.0 H 1B 40±1 local ~2.3GB 131k LM Studio › 620 Llama3-OpenBioLLM-8B 40±2 local ~6.6GB llama.cpp › 621 Nous Hermes 2 Mistral 7B … 40±2 local ~6.1GB 33k llama.cpp › 622 Nous Hermes 2 Mixtral 8x7… 40±2 local ~31GB 33k llama.cpp › 623 Deita 7B v1.0 40±2 local ~32GB how to run › 624 Code Llama 34B Instruct 40±1 local ~22GB 16k LM Studio › 625 Code Llama 34B (base) 40±1 local ~23GB 16k Ollama › 626 Starling-LM 7B beta 40±1 local ~6GB 8k LM Studio › 627 LFM2 1.2B 40±2 local ~2.2GB 33k LM Studio › 628 Code Tulu 2 13B 40±2† local ~29GB how to run › 629 LLaVA-v1.6-34B 40±2 local ~24GB 4k Ollama › 630 Bielik 1.5B v3.0 Instruct 40±3 local ~2.4GB LM Studio › 631 OpenHermes 2.5 Mistral 7B 39±1 local ~6.1GB llama.cpp › 632 Granite 3.3 2B Instruct 39±1† local ~3GB 128k LM Studio › 633 Phi-3-mini-128k-instruct … 39±2 local ~9.5GB how to run › 634 Mistral 7B Instruct v0.3 39±1 local ~5.8GB 32k LM Studio › 635 Vicuna 33B v1.3 39±1 local ~23GB 2k Ollama › 636 DeepSeek-Coder 6.7B Base 39±1 local ~5.8GB 16k llama.cpp › 637 Starling-LM 7B alpha 39±1 local ~6.1GB 8k llama.cpp › 638 Granite 8B Code Base (4K) 39±1 local ~6.6GB llama.cpp › 639 StarCoder2 15B 39±1 local ~11GB 16k LM Studio › 640 speechless-code-mistral-7… 39±1 local ~17GB how to run › 641 Falcon-H1 3B Instruct 39±4 local ~3.4GB 131k LM Studio › 642 Phi-2 (2.7B) 39±1 local ~3.2GB 2k LM Studio › 643 Granite 3.0 2B Instruct 39±1 local ~3.2GB 4k llama.cpp › 644 Llama 2 70B Chat 39±1 local ~32GB 4k llama.cpp › 645 Code Llama 13B Instruct 39±1 local ~9.8GB 16k llama.cpp › 646 MistralHermes-CodePro-7B-… 39±1 local ~6.1GB llama.cpp › 647 Llama-3-Instruct-8B-SimPO… 39±1 local ~6.6GB llama.cpp › 648 Llama 3.2 3B Instruct 38±1 api $0.12/M +local ~3.4GB 131k cloudflare.com ↗ 649 Qwen-14B-Chat 38±1 local ~11GB llama.cpp › 650 Falcon-180B-Chat 38±2 local ~115GB 2k Ollama › 651 CodeGemma 7B (v1.0) 38±1 local ~7.1GB llama.cpp › 652 Code Llama 13B Python 38±1 local ~9.8GB 16k llama.cpp › 653 Tulu 2 13B 38±2 local ~9.8GB llama.cpp › 654 Phi-3-mini-4k-instruct (A… 38±2 local ~4GB llama.cpp › 655 Apertus 8B Instruct (2509) 38±4 local ~6.3GB 66k LM Studio › 656 LLaMA 65B 38±1 comparison only 657 Granite 3B Code Instruct … 38±2 local ~3.7GB llama.cpp › 658 DeepSeek-V2 Lite Chat (16… 38±1 local ~11GB 32k LM Studio › 659 BioMistral-7B 38±3 local ~6.1GB llama.cpp › 660 Qwen3.5 0.8B 38±1 local ~2GB 262k llama.cpp › 661 Code Tulu 2 7B 38±2† local ~16GB how to run › 662 SOLAR 10.7B Instruct v1.0 37±1 local ~8.3GB 4k llama.cpp › 663 Phi-3.5-vision-instruct 37±2 local ~10GB how to run › 664 Tulu 2 DPO 13B 37±1 local ~9.8GB llama.cpp › 665 Yi-1.5 6B (base) 37±3 local ~5.1GB 4k LM Studio › 666 Qwen2-VL 2B Instruct 37±1 local ~2.8GB 33k LM Studio › 667 DeepSeek R1 Distill Qwen … 37±2 local ~2.6GB 131k LM Studio › 668 Gemma 2 2B IT 37±1 local ~3.1GB 8k LM Studio › 669 Qwen2.5 Coder 0.5B Instru… 37±1 local ~1.8GB 33k LM Studio › 670 Granite 20B Code Base (8K) 37±1 local ~15GB llama.cpp › 671 Gemma 1.1 7B IT 37±1 local ~7.1GB 8k llama.cpp › 672 Code Llama 7B Python 37±1 local ~5.8GB 16k llama.cpp › 673 Dolphin 2.2.1 Mistral 7B 37±2 local ~6.1GB llama.cpp › 674 Code Llama 13B (base) 37±1 local ~9.2GB 16k LM Studio › 675 Qwen1.5 7B Chat 37±1 local ~6.5GB 32k llama.cpp › 676 Yi 1.5 6B Chat 37±2 local ~5.1GB 4k LM Studio › 677 Falcon3 3B Instruct 37±3 local ~3.4GB 32k LM Studio › 678 Zephyr 7B alpha 36±1 local ~6.1GB llama.cpp › 679 Vicuna 13B v1.5 36±2 local ~9.8GB 4k llama.cpp › 680 Granite 3.1 2B Instruct 36±1 local ~3.1GB 128k llama.cpp › 681 EXAONE 3.5 2.4B Instruct 36±2 local ~3GB 33k LM Studio › 682 LLaMA 33B 36±1 comparison only 683 Code Llama 7B Instruct 36±1 local ~5.5GB 16k LM Studio › 684 LFM2 700M 36±2 local ~1.9GB 33k LM Studio › 685 Granite 3.2 2B Instruct 36±1 local ~3GB 131k LM Studio › 686 Mistral 7B Instruct v0.2 36±1 local ~6GB 32k LM Studio › 687 WizardLM 13B V1.2 36±1 local ~9.8GB llama.cpp › 688 Tulu 2 7B 36±2 local ~5.8GB llama.cpp › 689 xDAN-L1-Chat-RL-v1 36±1 local ~6.1GB llama.cpp › 690 Llama 3 8B (base) 36±1 local ~6.6GB llama.cpp › 691 Mistral 7B v0.1 (base) 36±0 local ~5.8GB 8k LM Studio › 692 Qwen2.5 1.5B Instruct 36±1 local ~2.4GB 33k LM Studio › 693 MPT-30B-Chat 35±2 comparison only 694 Gemma 7B (base) 35±1 local ~7.1GB 8k llama.cpp › 695 Tulu 2 DPO 7B 35±1 local ~5.8GB llama.cpp › 696 Eurus-2-7B-SFT 35±3† local ~6.4GB llama.cpp › 697 Falcon 40B (base) 35±1 local ~27GB 2k Ollama › 698 StarCoder2 7B 35±1 local ~6.1GB 16k LM Studio › 699 Granite 3.0 3B-A800M Inst… 35±1 local ~3.7GB 4k llama.cpp › 700 DeepSeek LLM 7B Chat 35±2† local ~5.7GB Ollama › 701 StarCoder 15.5B 35±1 local ~11GB Ollama › 702 Granite 3.1 3B-A800M Inst… 35±2† local ~3.6GB 131k llama.cpp › 703 Llama 2 13B (base) 34±1 local ~9.3GB 4k Ollama › 704 Code Llama 7B (base) 34±1 local ~5.5GB 16k LM Studio › 705 Nous-Hermes-13b 34±2 local ~9.8GB 2k llama.cpp › 706 ChatGLM3 6B 128K 34±4 local ~5.8GB 131k llama.cpp › 707 Zephyr 7B beta 34±1 local ~6.1GB llama.cpp › 708 PaLM 2 Chat (chat-bison@0… 34±2 comparison only 709 Qwen3 0.6B 33±1 local ~1.8GB 33k LM Studio › 710 Llama 2 13B Chat 33±1 local ~9.8GB 4k llama.cpp › 711 Flan-T5 XXL 33±1 local ~49GB how to run › 712 StarCoder2 3B 33±1 local ~3.4GB 16k llama.cpp › 713 Granite 3B Code Base (2K) 33±1 local ~3.7GB llama.cpp › 714 GPT-3 davinci (API base m… 33±1 comparison only 715 StarCoderBase 15.5B 33±1 local ~12GB llama.cpp › 716 Vicuna 7B v1.5 33±1 local ~5.8GB 4k llama.cpp › 717 Guanaco 33B 33±2 local ~70GB how to run › 718 Stable Code 3B 33±1 local ~3.3GB 16k llama.cpp › 719 OLMo 2 1B Instruct (0425) 33±2 local ~2.4GB 4k llama.cpp › 720 DeepSeek-Coder 1.3B Base 32±1 local ~2.4GB 16k Ollama › 721 Orca 2 13B 32±2 local ~9.8GB 4k llama.cpp › 722 Gemma 7B IT 32±1 local ~7.1GB 8k llama.cpp › 723 Mistral 7B Instruct v0.1 32±1 local ~6.1GB 8k llama.cpp › 724 Qwen1.5 4B Chat 32±2 local ~4.1GB 32k llama.cpp › 725 Ghost 7B Alpha 32±2 local ~6.1GB llama.cpp › 726 CodeGemma 2B (v1.0) 32±1 local ~3.2GB llama.cpp › 727 Qwen-7B (base) 32±1 local ~6.6GB 33k llama.cpp › 728 Phi-1.5 (1.3B) 32±1 local ~2.4GB 2k llama.cpp › 729 Falcon-H1 1.5B Instruct 31±4 local ~2.4GB 131k LM Studio › 730 Flan-T5 XL 31±2 local ~14GB how to run › 731 Qwen2 1.5B Instruct 31±1 local ~2.4GB 32k LM Studio › 732 Qwen-7B-Chat 31±1 local ~6.6GB 33k llama.cpp › 733 Gemma 3 1B 31±1 local ~2.2GB 32k LM Studio › 734 StarCoderBase 7B 31±1 local ~6GB Ollama › 735 Llama 2 7B (base) 31±1 local ~5.5GB 4k Ollama › 736 OPT-66B 31±1 local ~141GB 2k how to run › 737 LFM2 350M 30±2 local ~1.7GB 33k LM Studio › 738 SmolLM2 1.7B Instruct 30±1 local ~2.6GB 8k LM Studio › 739 Llama 3.2 1B Instruct 30±1 api $0.020/M +loc… ~2.2GB 131k novita.ai ↗ 740 Falcon 7B (base) 30±1 local ~5.9GB 2k Ollama › 741 Llama 2 7B Chat 30±1 local ~5.8GB 4k llama.cpp › 742 Baichuan2-7B-Base 30±1 local ~17GB how to run › 743 LLaMA 7B 30±1 comparison only 744 MPT-7B (base) 30±1 comparison only 745 OLMo 7B Instruct 29±2 local ~16GB how to run › 746 StableLM-3B-4E1T 29±1 local ~3.3GB 4k llama.cpp › 747 Gemma 2B (base) 29±1 local ~3.2GB 8k llama.cpp › 748 OpenLLaMA 7B v2 29±1 local ~9.1GB llama.cpp › 749 Gemma 1.1 2B IT 29±1 local ~3.2GB 8k llama.cpp › 750 StableLM-Base-Alpha 7B v2 29±1 local ~30GB 4k how to run › 751 Granite 3.1 1B-A400M Inst… 29±2† local ~2.3GB 131k llama.cpp › 752 OPT-30B 29±1 local ~64GB 2k how to run › 753 NexusRaven V2 13B 29±3 local ~9.3GB 16k Ollama › 754 OLMo 7B 28±1† local ~30GB 2k how to run › 755 Falcon 40B Instruct 28±2 local ~27GB Ollama › 756 GPT-NeoX-20B 28±1 local ~14GB 2k LM Studio › 757 LLaMA 13B 28±1 comparison only 758 Koala 13B 28±1 comparison only 759 StarCoderBase 3B 28±1 local ~3.4GB Ollama › 760 Granite 3.0 1B-A400M Inst… 27±1 local ~2.3GB 4k llama.cpp › 761 RedPajama-INCITE 7B Base 27±1 local ~16GB 2k how to run › 762 ChatGLM3-6B 27±2 local ~5.8GB llama.cpp › 763 Granite 4.0 H 350M 27±1† local ~1.7GB 33k LM Studio › 764 Flan-T5 Large 27±2 local ~4.8GB how to run › 765 MPT-7B-Chat 27±2 comparison only 766 Gemma 2B IT 27±1 local ~3.1GB 8k llama.cpp › 767 EuroLLM 1.7B Instruct 27±3 local ~2.4GB 4k LM Studio › 768 Qwen2.5 0.5B Instruct 27±2 local ~1.8GB 33k LM Studio › 769 GPT-3 text-davinci-001 26±3 comparison only 770 GPT4All-13B-Snoozy 26±2 local ~56GB how to run › 771 OPT-13B (13 billion param… 26±1 local ~29GB 2k how to run › 772 Granite 4.0 350M 26±1 local ~1.7GB 32k LM Studio › 773 GPT-3 curie (API base mod… 26±1 comparison only 774 Alpaca 13B 26±2 comparison only 775 Falcon 7B Instruct 25±2 local ~5.9GB Ollama › 776 Qwen2 0.5B Instruct 25±2 local ~1.8GB 32k LM Studio › 777 Pythia 12B 25±1 local ~9.5GB 2k llama.cpp › 778 RedPajama-INCITE Base 3B … 25±3 local ~3.4GB 2k llama.cpp › 779 ChatGLM2-6B 25±2 local ~15GB how to run › 780 GPT-J 6B 25±1 local ~14GB 2k how to run › 781 OpenLLaMA 3B v2 25±1 local ~4.2GB llama.cpp › 782 OPT-6.7B 24±1 local ~16GB 2k how to run › 783 RWKV-4-Raven 14B 24±2 comparison only 784 Falcon3 1B Instruct 24±4 local ~2.4GB 8k LM Studio › 785 StableLM-Base-Alpha 3B v2 24±1 local ~13GB 4k how to run › 786 Qwen1.5 1.8B Chat 24±2 local ~2.7GB 33k Ollama › 787 Pythia 6.9B 23±1 local ~6.1GB 2k llama.cpp › 788 OpenAssistant SFT-4 Pythi… 22±1 local ~9.5GB llama.cpp › 789 Pythia 2.8B 21±1 local ~3.4GB 2k llama.cpp › 790 BLOOM 7.1B 21±1 local ~7.1GB 2k llama.cpp › 791 OLMo 1B 21±2† local ~2.2GB 2k llama.cpp › 792 Phi-1 (1.3B) 21±4 local ~2.4GB 2k llama.cpp › 793 Qwen1.5 0.5B Chat 21±3 local ~1.8GB 33k LM Studio › 794 ChatGLM-6B 20±2 local ~16GB how to run › 795 Pythia 2.8B (deduped) 20±1 local ~3.4GB 2k llama.cpp › 796 Flan-T5 Base 20±3† local ~2.5GB how to run › 797 Falcon-RW 1B 20±2 local ~4.3GB 2k how to run › 798 OPT-2.7B 19±1 local ~7.1GB 2k how to run › 799 SmolLM2 360M Instruct 19±3 local ~1.7GB 8k LM Studio › 800 GPT-Neo 2.7B 18±1 local ~13GB 2k how to run › 801 GPT-3 babbage (API base m… 18±1 comparison only 802 Pythia 1.4B 18±2 local ~2.4GB 2k llama.cpp › 803 BLOOM 3B 18±1 local ~4GB 2k llama.cpp › 804 FastChat-T5 3B 17±2 local ~8.5GB how to run › 805 Dolly v2 12B 17±2 comparison only 806 OPT-1.3B 16±1 local ~4.3GB 2k how to run › 807 Gemma 3 270M IT 15±4 local ~1.7GB 32k LM Studio › 808 Pythia 1B 14±1 local ~2.2GB 2k llama.cpp › 809 StableLM-Tuned-Alpha 7B 14±2 local ~6.9GB llama.cpp › 810 BLOOM 1.7B 14±1 local ~3.1GB 2k llama.cpp › 811 Flan-T5 Small 14±3† local ~1.8GB how to run › 812 GPT-Neo 1.3B 14±2 local ~7.1GB 2k how to run › 813 GPT-2 XL (1.5B) 11±1 local ~2.8GB 1k llama.cpp › 814 BLOOM 1.1B 11±1 local ~2.6GB 2k llama.cpp › 815 Pythia 410M 10±2 local ~1.8GB 2k llama.cpp › 816 GPT-3 ada (API base model) 10±2 comparison only 817 StableLM-Base-Alpha 7B 9±1 local ~35GB 4k how to run › 818 Zyte-1B 8±2 local ~2.2GB llama.cpp › 819 GPT-Neo 125M 5±5 local ~2.1GB 2k how to run › 820 BLOOM 560M 5±2 local ~2.1GB 2k llama.cpp › 821 OPT-350M 5±2 local ~2.2GB 2k how to run › 822 StableLM-Base-Alpha 3B 4±1 local ~17GB 4k how to run › 823 Pythia 160M 3±2 local ~1.6GB 2k llama.cpp › 824 OPT-125M 3±2 local ~1.8GB 2k how to run › 825 GPT-2 Large (774M) 1±2 local ~2GB 1k llama.cpp › 826 GPT-2 Medium (355M) -1±2 local ~1.8GB 1k llama.cpp › 827 Pythia 70M -7±2 local ~1.6GB 2k llama.cpp › 828 GPT-2 Small (124M) -9±3 local ~1.6GB 1k LM Studio › — Grok Build 0.1 no reviewed results api $1.25/M 256k x.ai ↗ — K2 Horizon 7B not scored yet local ~7.7GB 524k llama.cpp ›
+ 304 models with fewer than 3 results aren't ranked here · show them
* new or thinly tested: ranked on too few benchmark results · † only the maker's own numbers so far · ~RAM is estimated; full context may need more memory · 830 of 1134 catalogue models listed here · sources and licences
whatmodel is an open-source AI leaderboard database. The data, scoring and ranking engine,
and benchmark tools are available here. This README and pages/ are generated from the
same data and code with npm run pages.
An interactive version of the database can be found at whatmodel.sh. Every model has its own page showing what it's good at, how each test result compares with other models, and how to run it:
| Path | Contents |
|---|---|
data/ | Models, benchmark results, prices, measured speeds and hardware. Every number links to its source. |
data/extracts/ | Share-alike source extracts, published separately as their licences require |
lib/ | The engine: scores (score.ts), memory fit and ranking (fit.ts), page data (views.ts), terminal output (format.ts), markdown pages (markdown.ts) |
pages/ | The other leaderboards, the method and one page per model, generated |
bench.sh, bench.py | Interactive benchmark launcher and runner; see BENCHMARKING.md |
scripts/ | Checks, import, export, page generation and measurement scripts |
licenses/ | Full licence notices for every source |
docs/ | Source audit, validation and coverage reports |
curl "whatmodel.sh/?ram=32&gpu=apple"
curl "whatmodel.sh/image?ram=64&gpu=card&vram=24&format=json"
| Parameter | Meaning |
|---|---|
ram, vram | GB, or raw bytes / kB / MiB as the OS reports them |
gpu | apple, card (NVIDIA, AMD or Intel graphics card) or none |
mac | bytes, optionally followed by a newline and the chip name; implies Apple Silicon |
nv | nvidia-smi --query-gpu=name,memory.total --format=csv,noheader output |
hw | exact chip or card id from data/hardware.json |
max | highest price in the category's unit (LLMs: $ per 1M tokens, blended 3:1 in:out); 0 = local only |
sort, dir | score, where, speed, ram or context; asc or desc |
set | optional model-set filter; omitted = all sets, without changing the overall score scale |
score | set uses only the selected set’s evaluations; omitted = overall scores, with uncalibrated scores/ranks reported as null |
all=1, n=20, format=json, color=1, chart=1 | show models that don't fit, row limit (text: 10, JSON: 40), JSON, ANSI colour, charts |
width=120, help=1 | wider terminal table (default: 80 columns), concise command help |
The same output runs from this repo, with no server:
npm install
npm run board -- "ram=32&gpu=apple"
npm run board -- image "ram=64&gpu=card&vram=24"
Each category's score comes from a one-factor latent model fitted to the benchmark results, within sets
of models that share enough benchmarks to be compared. It lives in lib/score.ts and is explained in
pages/method.md. 50 is an average model, and ± is the uncertainty. npm run check
reports how well the fit predicts results it hasn't seen.
Sponsors never change rankings or scores. Sponsored spots on whatmodel.sh are always labelled, and you can check the ranking against the code here.
data/ directly, or run npm run export to write cleaned snapshots, CSV tables
(scores, models, results, speeds and bench runs) and a SHA-256 manifest to dist/data/. They're the same
files whatmodel.sh serves at whatmodel.sh/data/manifest.json.data/sources.json lists them and
licenses/ has the full notices.The bench measures how fast models run on your machine. It finds the models and AI tools you have and the whatmodel.sh models that fit your memory; you tick what to test, confirm the plan, then review the measurements. It installs or downloads nothing without asking, and offers to set up Python if it's missing. Works on macOS and Linux; on Windows, run it in WSL.
curl -fsSL https://whatmodel.sh/bench.sh | sh
Results stay local until you choose to open a prefilled GitHub issue. Review it and submit it yourself in the browser. API and agent tests may incur provider charges. See BENCHMARKING.md for supported runners, measurement protocols and what the public report contains.
Submitted results are released under CC0 1.0 and reviewed before they're added.
Corrections and new results are welcome, and you don't need to write code. See CONTRIBUTING.md.
Code: MIT. Data: CC BY-SA 4.0. Bench results: CC0 1.0. Copyright 2026 Merlinsbeard Pty Ltd.
Find the best AI models you can run
See the codeThe open AI model leaderboard. Every model is ranked by one score built from 19,731 published benchmark results, with its price, the memory it needs and the command to run it.
Filter by your computer or budget at whatmodel.sh, or read the same leaderboard below. About this repo
llm · image · video · tts · stt
updated 2026-10-11 02:20 UTC · 830 models · scores combine 472 benchmarks · how it works · benchmark your machine · interactive version
Run curl "whatmodel.sh/?mac=$(sysctl -n hw.memsize)" to see what your Mac can run, or curl whatmodel.sh for this list in your terminal.
curl "whatmodel.sh/?ram=$(awk '/MemT/{print $2}' /proc/meminfo)&vram=$(nvidia-smi -i 0 --query-gpu=memory.total --format=csv,noheader,nounits 2>/dev/null || cat /sys/class/drm/card*/device/mem_info_vram_total 2>/dev/null | sort -n | tail -1)" # Linux
curl.exe "whatmodel.sh/?ram=$((gcim Win32_ComputerSystem).TotalPhysicalMemory)&vram=$(nvidia-smi -i 0 --query-gpu=memory.total --format=csv,noheader,nounits 2>$null)" # Windows
16GB Mac · 32GB Mac · 64GB Mac · 128GB Mac · 12GB graphics card · 24GB graphics card · 16GB, no graphics card
score vs API price
│ MiMo-V2.6-Pro GPT-6.1 Sol ◆──────◆────────◆ GPT-6 Astra
75 ┤ · · · · · · · · ●─●──∘─●◆∘∘●∘───◇∘●─∘─∘◇∘∘∘∘∘∘◇◇∘──◇◇◇◇◇◇◇◇ ·◇·◇◇ · · ◇ · ·◇·◇· ◇ ◇ ·
│ ◆─∘──∘∘∘∘∘∘ ◇∘∘ ◇∘∘∘∘∘ ∘◇∘∘∘◇∘∘∘ ∘◇◇ ∘ ∘◇◇∘◇◇ ◇ ◇ ◇
50 ┤ · · · · · ●┘◇∘∘ ·∘◇∘∘ · ·∘·∘∘ · · · ·◇· · ·◇· · · · · ·∘· · · ··· · · · · · ◇ · · · ·
│ ●────────┘
└──────────────────┬──────────────────────┬──────────────────────┬─────────────────────
$0.1 $1 $10
score vs API price (USD per 1M tokens, 3 input : 1 output, log)
● open weights ◆ API only ─┘ best for the price
│ Claude Opus 5.5
│ · · ◆────────◆ GPT-6 Astra
│ GPT-6.1 Sol ◆──────┘ ◇ ◇
80 ┤ · · · · · · · · ··· MiMo-V2.6-Pro ●───────∘───────────┘ · · ·◇··◇ · · ◇ · · · · · ◇ ·
│ ●◆─◇●────◇─┘ ◇∘ ◇∘ ◇ ◇ ◇∘ ◇ ◇
│ ●─●────┘ ∘ ◇ ·◇ ∘◇◇ ∘ ◇◇◇◇ ◇ ◇ ◇ ◇
│ │ ∘ ∘ ∘∘ ◇∘∘ ◇ ∘◇∘∘∘∘∘∘◇◇∘ ◇ ◇ ◇
│ ◆────┘· ∘ ◇∘ ∘ ∘ ∘ ∘∘∘ ∘∘∘ ◇◇ ◇ ∘◇ · ◇
60 ┤ · · · · · ·│· · ∘ ∘ ◇∘· · ·∘∘∘·∘∘ · ◇∘◇∘∘ ∘ · · · · ◇◇∘ · · · · · · · · · · · · · · ·
│ ● ∘ ∘·∘∘∘ ◇ ∘ ∘ ∘ ∘ ◇∘∘∘◇· ◇ ◇ ·
│ │ ∘∘∘ ∘ ◇∘ ◇∘ ∘ ∘ ∘ ∘ ◇ ◇
│ │ ∘ ·◇ ∘ ∘ · ◇ · ◇
│ ●┘◇ ∘ ∘ ∘ ∘ ◇ ∘
│ │ ∘ · ∘ · ·
40 ┤ · · · · · │ · · · ·∘· · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · ·
│ │ · · ·
│ ●─Llama 3.2 1B Instruct
│ · · ·
└──────────────────┬──────────────────────┬──────────────────────┬─────────────────────
$0.1 $1 $10
score vs memory to run it (GB, log)
─┘ best for the memory
│ Hy4 preview (770B-A49B)
│ · MiMo-V2.6-Flash (309B-A15… ●────∘● MiMo-V2.6-Pro
75 ┤ · · · · · · · ·Qwen3.8 27B·●────────────●────∘∘●─∘●─────∘┘· ∘∘· · ∘∘· ·∘·∘· · · · ∘ ·
│ Spark-X2.5 4B ∘∘∘ ∘ ∘∘ ∘ ∘∘∘∘ ∘∘∘ ∘ ∘∘ ∘ ∘∘ · ∘
│ ●──●────∘─∘∘─∘∘─────∘∘∘∘∘∘ ∘ ∘ ∘∘∘ ∘∘∘ ∘∘ ∘∘ ∘∘ ∘ ∘
│ ∘ ∘ ∘ ∘∘ ∘∘ ∘∘∘∘∘∘∘ ∘∘∘∘∘ ∘ ∘ ∘∘ ∘ ∘∘ ∘ ∘ ·
50 ┤ ·●────∘ ·∘· · ∘ ∘∘·∘∘ · ·∘∘∘∘∘∘∘∘∘∘ ·∘· ∘ ∘ ·∘· ∘ ·∘· · · ·∘· · · · · ·∘· · · · · · ·
│ ∘ ∘ ∘ ∘ ∘ ∘∘ ∘∘∘∘ ∘ ∘ ∘∘ ∘∘∘∘ ∘∘ ∘∘∘ ∘∘ · ∘ ·
│ ∘ ∘ ∘ ∘ ∘ ∘∘ ∘∘∘∘ ∘∘∘ ∘∘ ∘∘∘ ∘ ∘ ∘
│ ∘ ∘ ∘ ∘ ∘ ∘ ∘∘∘∘∘ ∘ ∘ ∘ · ·
25 ┤ ·∘· · ∘ · · ∘ ∘ · ·∘· · ∘∘∘ · ·∘∘∘· · · ∘∘· · · · ∘ · · · · · · · · · · · · · · · · ·
│ ∘ ∘ ∘ ∘ ∘ ∘· ∘∘∘ · ·
│ ∘ ∘ ∘ ∘ ∘ ∘
│ ∘ ∘ · ∘ · ·
│ ∘ ∘
0 ┤ ·∘· · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · ··· · · · · ·
│ ∘
│ · · ·
└─────────────────────┬─────────────────────────┬──────────────────────────┬───────────
10GB 100GB 1TB
# model score where RAM context get it 1 GPT-6 Astra 87±1 api $20.00/M 1.1M openai.com ↗ 2 Claude Opus 5.5 87±1 api $8.00/M 1M anthropic.com ↗ 3 GPT-6.1 Sol 85±1 api $4.00/M 1.1M openai.com ↗ 4 Claude Fable 5.1 85±1 api $20.00/M 1M anthropic.com ↗ 5 Claude Sonnet 5.5 84±1 api $4.00/M 1M claude.com ↗ 6 Claude Opus 5 83±1 api $10.00/M 1M anthropic.com ↗ 7 GPT-6 Sol 83±1 api $4.00/M 1.1M openai.com ↗ 8 GPT-5.5 Pro 82±1 api $67.50/M 1.1M openai.com ↗ 9 Gemini 4 Argon (limited p… 82±1 limited preview 10 GPT-5.6 Sol 82±1 api $8.00/M 1.1M openai.com ↗ 11 Claude Fable 5 81±1 api $20.00/M 1M anthropic.com ↗ 12 Claude Mythos Preview (li… 81±1† limited preview 1M 13 GPT-5.4 Pro 81±1 api $67.50/M 1.1M openai.com ↗ 14 GPT-5.5 80±1 api $11.25/M 1.1M openai.com ↗ 15 MiMo-V2.6-Pro 79±1 api $0.54/M +local ~443GB 1M mimo.mi.com ↗ 16 Hy4 preview (770B-A49B) 79±1 api $1.25/M +local ~248GB 1M tencent.com ↗ 17 GPT-5.6 Terra 79±1 api $4.50/M 1.1M openai.com ↗ 18 Qwen3.8 Max (2.4T-A95B) 78±1 api $3.00/M +local ~380GB 262k deepinfra.com ↗ 19 Claude Opus 4.8 78±0 api $10.00/M 1M anthropic.com ↗ 20 GPT-5.4 78±1 api $5.63/M 1.1M openai.com ↗ 21 DeepSeek V4.1 Flash 78±1 api $0.26/M +local ~537GB 1M deepseek.com ↗ 22 Claude Haiku 5.5 77±1 api $0.20/M 1M claude.com ↗ 23 Gemini 3.8 Flash 77±1 api $1.50/M 1M google.dev ↗ 24 Muse Spark 1.3 77±1 api $2.00/M 1M dev.meta.ai ↗ 25 Kimi K3 77±1 api $6.00/M +local ~884GB 1M moonshot.ai ↗ 26 GPT-5.2 Pro 77±1 api $57.75/M 400k openai.com ↗ 27 MiMo-V2.6-Flash (309B-A15… 77±1 api $0.17/M +local ~136GB 1M mimo.mi.com ↗ 28 GPT-6 Luna 77±1 api $0.20/M 1.1M openai.com ↗ 29 GPT-5.6 Luna 77±1 api $0.45/M 1.1M openai.com ↗ 30 Grok 4.6 77±1 api $3.00/M 500k x.ai ↗ 31 DeepSeek V4 Pro (0813) 76±1 api $1.63/M +local ~1010GB 1M deepinfra.com ↗ 32 Gemini 3.7 Flash 76±1 api $1.50/M 1M google.dev ↗ 33 GLM-5.3 76±1 api $2.15/M +local ~239GB 1M z.ai ↗ 34 Qwen3.8 Max (0902) 76±1 api $3.00/M 1M alibabacloud.co… 35 DeepSeek V3.2 Speciale 76±3† comparison only 36 DeepSeek V4 Flash Vision … 76±2 local ~104GB 1M llama.cpp › 37 Qwen3.8-Flash-Next 75±1 local ~59GB 262k llama.cpp › 38 Qwen3.8-Omni-Flash 75±2† api $0.23/M 1M alibabacloud.co… 39 Gemini 3.1 Pro (Preview) 75±1 api $4.50/M 1M google.dev ↗ 40 Muse Spark 1.2 75±1 api $2.00/M 1M dev.meta.ai ↗ 41 Claude Opus 4.7 75±1 api $10.00/M 1M anthropic.com ↗ 42 Grok 4.5 75±1 api $3.00/M 500k x.ai ↗ 43 Muse Spark 1.1 75±1 api $2.00/M 1M dev.meta.ai ↗ 44 Muse Spark 75±2 comparison only 45 Step 5 Preview 75±1 comparison only 46 Claude Sonnet 5 75±1 api $4.00/M 1M anthropic.com ↗ 47 Grok 4.7 75±1 api $3.00/M 500k x.ai ↗ 48 Claude Opus 4.6 75±1 api $10.00/M 1M anthropic.com ↗ 49 GPT-5.3-Codex 74±1 api $4.81/M 400k openai.com ↗ 50 DeepSeek V4 Flash (0731) 74±1 api $0.11/M +local ~92GB 1M deepinfra.com ↗ 51 Qwen3.7 Max 74±1 api $3.75/M 1M deepinfra.com ↗ 52 GLM-5.3-Flash 74±1 api $0.24/M +local ~103GB 1M z.ai ↗ 53 Gemini 3.5 Flash 74±1 api $3.38/M 1M google.dev ↗ 54 GPT-5.2 74±1 api $4.81/M 400k openai.com ↗ 55 Hunyuan Hy3 (295B-A21B) 74±1 api $0.23/M +local ~96GB 256k deepinfra.com ↗ 56 Gemini 3.6 Flash 74±1 api $1.50/M 1M google.dev ↗ 57 GLM-5.2 74±1 api $2.15/M +local ~121GB 1M z.ai ↗ 58 Dola Seed 2.0 Pro 74±2 comparison only 59 Qwen3.8 27B 73±1 api $1.60/M +local ~18GB 262k groq.com ↗ 60 Grok 4.20 Multi-Agent (03… 73±2 api $1.56/M 1M x.ai ↗ 61 Kimi K2.6 73±1 api $1.71/M +local ~317GB 262k moonshot.ai ↗ 62 GPT-5 Pro 73±1 api $41.25/M 400k openai.com ↗ 63 LongCat 2.0 (1.6T-A48B) 73±2 api $1.30/M +local ~2155GB 1M longcat.chat ↗ 64 Gemini 3 Pro (Preview) 73±1 comparison only 1M 65 GPT-5 72±1 api $3.44/M 400k openai.com ↗ 66 Qwen3.6 Max Preview 72±1 api $2.92/M 256k alibabacloud.co… 67 Claude Sonnet 4.6 72±1 api $6.00/M 1M anthropic.com ↗ 68 Gemini 3 Flash (Preview) 72±1 api $1.13/M 1M google.dev ↗ 69 DeepSeek V4 Pro (Preview,… 72±1 local ~604GB llama.cpp › 70 Claude Opus 4.5 72±1 api $10.00/M 200k anthropic.com ↗ 71 Qwen3.7 Plus 72±1 api $0.70/M 1M alibabacloud.co… 72 Kimi K2.7 Code 72±1 api $1.71/M +local ~359GB 262k kimi.ai ↗ 73 DeepSeek V4 Flash (Previe… 72±1 local ~103GB llama.cpp › 74 Grok 4.20 (0309 reasoning) 72±1 api $1.56/M 1M x.ai ↗ 75 Solar Open 2 (250B-A15B) 72±1† local ~527GB 1M how to run › 76 North Small Translate 1.0 71±6† api price unknown 16k cohere.com ↗ 77 GPT-5.4 mini 71±1 api $1.69/M 400k openai.com ↗ 78 Ling 3.0 Flash (124B-A5B) 71±2† api $0.090/M +loc… ~75GB 131k deepinfra.com ↗ 79 Mistral Large 4 (public p… 71±2 api $2.06/M 1M mistral.ai ↗ 80 GPT-5.1 71±1 api $3.44/M 400k openai.com ↗ 81 GLM-5.1 71±1 api $2.15/M +local ~119GB 200k z.ai ↗ 82 Qwen3.5 397B-A17B 71±0 api $1.09/M +local ~65GB 262k deepinfra.com ↗ 83 Ornith 1.5 35B-A3B 71±2† local ~22GB 262k LM Studio › 84 MiniMax-M3 70±1 api $0.48/M +local ~137GB 1M deepinfra.com ↗ 85 Grok 4.3 70±1 api $1.56/M 1M x.ai ↗ 86 Gemini 2.5 Pro Preview (0… 70±2 comparison only 87 o3-pro 70±2 api $35.00/M 200k openai.com ↗ 88 KAT-Coder-V2.5-Dev (35B-A… 70±2† local ~22GB 262k LM Studio › 89 Qwen3.6 Plus 70±1 api $1.13/M 1M together.ai ↗ 90 Inkling Small (266B-A12B) 70±1 api $0.64/M +local ~86GB 1M deepinfra.com ↗ 91 MiMo-V2-Pro 70±1 comparison only 92 Kimi K2.5 70±1 api $1.20/M +local ~160GB 262k aws.amazon.com ↗ 93 GLM-5 70±1 api $1.55/M +local ~119GB 205k z.ai ↗ 94 Qwen3-Max (2026-01-23) 70±1† comparison only 95 MiMo-V2.5-Pro (1T-A42B) 70±1 api $0.65/M +local ~163GB 1M novita.ai ↗ 96 Inkling (975B-A41B) 70±1 api $1.73/M +local ~156GB 1M deepinfra.com ↗ 97 Muse Glimmer 30B 70±1 api $0.64/M +local ~19GB 131k fireworks.ai ↗ 98 Nemotron 3 Ultra 550B-A55B 69±1 api $1.35/M +local ~89GB 1M together.ai ↗ 99 Qwen3.6 27B 69±1 api $1.20/M +local ~18GB 262k groq.com ↗ 100 Ring-2.6-1T 69±2† comparison only 101 A.X K2 (688B-A33B) 69±1† local ~1454GB 262k how to run › 102 o3 69±1 api $3.50/M 200k openai.com ↗ 103 MiMo-V2.5 (310B-A15B) 69±1 api $0.21/M +local ~100GB 1M novita.ai ↗ 104 Qwen3.5 Plus 69±2 api $0.90/M 1M alibabacloud.co… 105 Qwen3.5 122B-A10B 69±1 api $1.10/M +local ~75GB 262k alibabacloud.co… 106 Gemini 3.5 Flash-Lite 69±1 api $0.85/M 1M google.dev ↗ 107 GLM-4.7 69±1 api $1.00/M +local ~58GB 205k z.ai ↗ 108 GPT-5.5 Instant 69±1 comparison only 109 GPT-5.4 nano 69±1 api $0.46/M 400k openai.com ↗ 110 Claude Sonnet 4.5 69±1 api $6.00/M 200k anthropic.com ↗ 111 Qwen3.5 27B 69±0 api $0.84/M +local ~18GB 262k deepinfra.com ↗ 112 GLM-5V-Turbo 69±2 api $1.90/M 200k siliconflow.com… 113 GPT-5.2-Codex 69±2 comparison only 114 LongCat-Flash-Thinking (5… 69±1† comparison only 128k 115 Step 3.5 Flash (196B-A11B) 68±1 api $0.15/M +local ~118GB 256k stepfun.ai ↗ 116 DeepSeek V3.2 68±1 api $0.29/M +local ~110GB 128k deepinfra.com ↗ 117 MiMo-V2-Omni 68±2 comparison only 262k 118 Gemini 2.5 Pro Experiment… 68±3 comparison only 119 Grok 4 (0709) 68±1 comparison only 120 MiniMax-M2.7 68±1 api $0.53/M +local ~75GB 205k minimax.io ↗ 121 Qwen3.6 35B-A3B 68±1 api $0.31/M +local ~23GB 262k deepinfra.com ↗ 122 Ornith 1.0 35B (35B-A3B) 68±1 local ~23GB 262k LM Studio › 123 Hy3 preview (295B-A21B) 68±1 local ~629GB how to run › 124 GPT-5 mini 67±1 api $0.69/M 400k openai.com ↗ 125 Gemini 2.5 Pro 67±1 api $3.44/M 1M google.dev ↗ 126 Claude Opus 4 67±1 comparison only 127 K-EXAONE 2.0 (750B-A37B) 67±1† local ~791GB 262k how to run › 128 Claude Opus 4.1 67±1 comparison only 129 Kimi K2 Thinking 67±1 api $1.07/M +local ~160GB 262k aws.amazon.com ↗ 130 Laguna M.1 67±2 api price unknown… ~135GB 262k huggingface.co ↗ 131 Qwen3.7 Flash 67±1 api $0.055/M 1M alibabacloud.co… 132 Step 3.7 Flash (198B-A11B) 67±1 api $0.44/M +local ~104GB 256k stepfun.ai ↗ 133 Gemma 4 31B 67±1 api $0.19/M +local ~21GB 262k deepinfra.com ↗ 134 Gemini 3.1 Flash-Lite 67±1 api $0.56/M 1M google.dev ↗ 135 Qwen3-Coder-Next (80B-A3B) 67±2 api $0.60/M +local ~27GB 262k alibabacloud.co… 136 Gemini 2.5 Pro Preview (0… 67±2 comparison only 137 Gemini 2.5 Pro Preview (0… 67±1 comparison only 138 GigaChat 3.5 Ultra Reason… 67±1† local ~461GB 262k how to run › 139 Qwen3.5 35B-A3B 67±0 api $0.35/M +local ~23GB 262k deepinfra.com ↗ 140 DeepSeek V3.1 Terminus 67±1 local ~230GB 164k llama.cpp › 141 DeepSeek V3.2 Exp 67±1 local ~726GB how to run › 142 Qwen3 235B-A22B Thinking … 66±1 api $0.97/M +local ~76GB 131k novita.ai ↗ 143 Grok 4.1 Fast (Reasoning) 66±1 api $0.28/M 128k azure.microsoft… 144 o4-mini 66±1 api $1.93/M 200k openai.com ↗ 145 Qwen3.6 Flash 66±1 api $0.56/M 1M alibabacloud.co… 146 MiMo-V2-Flash (309B-A15B) 66±1 local ~108GB llama.cpp › 147 GLM-4.6 66±1 api $1.00/M +local ~58GB 205k z.ai ↗ 148 Kimi K2 Instruct (0711) 66±1 api $1.00/M +local ~160GB 131k novita.ai ↗ 149 o1 (2024-12-17) 66±1 api $26.25/M 200k openai.com ↗ 150 MiniMax-M2.5 66±1 api $0.53/M +local ~75GB 205k minimax.io ↗ 151 Qwen3-Max (2025-09-23) 66±1 api $2.40/M 262k alibabacloud.co… 152 DeepSeek V3.1 66±1 api $0.42/M +local ~110GB 131k deepinfra.com ↗ 153 Spark-X2.5 4B 66±1† local ~3.9GB 1M LM Studio › 154 MiniMax-M2.1 66±1 local ~79GB llama.cpp › 155 Nemotron Cascade 2 30B-A3B 66±1 local ~20GB 256k LM Studio › 156 Laguna XS 2.1 (33B-A3B) 66±2† local ~21GB 262k LM Studio › 157 LongCat-Flash-Chat (560B-… 66±1 local ~333GB 128k LM Studio › 158 o3-mini 65±1 api $1.93/M 200k openai.com ↗ 159 Claude 3.7 Sonnet 65±1 comparison only 160 Claude Sonnet 4 65±1 comparison only 161 MiMo-V2.6-Distill-Qwen-9B 65±2† local ~7.8GB LM Studio › 162 K-EXAONE (236B-A23B) 65±1 local ~152GB 262k llama.cpp › 163 DeepSeek R1 0528 65±1 api $0.91/M +local ~110GB 164k deepinfra.com ↗ 164 EXAONE 4.5 33B 65±2† local ~24GB 262k llama.cpp › 165 Qwen3.5 Flash 65±1 api $0.17/M 1M alibabacloud.co… 166 ChatGPT-4o (2025-03-26) 65±2 comparison only 167 INTELLECT-3 65±1 local ~48GB llama.cpp › 168 Nemotron 3 Super 120B-A12B 65±1 api $0.28/M +local ~73GB 262k aws.amazon.com ↗ 169 GLM-4.5 65±1 api $1.00/M +local ~58GB 131k z.ai ↗ 170 Solar Pro 4 65±2 api $0.53/M 524k upstage.ai ↗ 171 MiniMax-M2 65±1 api $0.53/M +local ~75GB 205k aws.amazon.com ↗ 172 Laguna XS.2 65±2 local ~21GB 262k LM Studio › 173 Qwen3 235B-A22B Instruct … 65±1 api $0.20/M +local ~76GB 262k deepinfra.com ↗ 174 Mistral Medium 3.5 (128B) 65±1 api $3.00/M +local ~42GB 262k mistral.ai ↗ 175 Command A+ (05-2026) 64±1 api price unknown… ~71GB 128k cohere.com ↗ 176 Claude Haiku 4.5 64±1 api $2.00/M 200k anthropic.com ↗ 177 Nemotron-Labs-3 Puzzle 75… 64±1† local ~35GB llama.cpp › 178 Qwen3-VL 235B-A22B Thinki… 64±1 local ~111GB 256k LM Studio › 179 Qwen3-Coder 480B-A35B Ins… 64±1 api $0.67/M +local ~78GB 262k novita.ai ↗ 180 Mistral Medium 3.1 (2508) 64±2 comparison only 181 Gemma 4 26B-A4B 64±1 api $0.14/M +local ~18GB 262k deepinfra.com ↗ 182 Kimi K2 Instruct (0905) 64±2 local ~298GB llama.cpp › 183 Qwen3-VL 235B-A22B Instru… 64±1 local ~134GB 256k LM Studio › 184 Nova 2 Pro (Preview) 64±2 api $3.78/M 1M aws.amazon.com ↗ 185 Grok 3 Mini Beta 64±1 comparison only 186 MiniCPM5 2B 64±1† local ~3GB 131k LM Studio › 187 Grok 3 Beta 63±1 comparison only 188 Qwen3-Next 80B-A3B Thinki… 63±1 api $0.41/M +local ~27GB 131k alibabacloud.co… 189 DeepSeek R1 (Jan '25) 63±1 api $4.00/M +local ~110GB 128k together.ai ↗ 190 Gemini 2.5 Flash Preview … 63±1 comparison only 191 Qwen3-Next 80B-A3B Instru… 63±1 api $0.34/M +local ~27GB 131k deepinfra.com ↗ 192 Granite 4.2 30B 63±1 api $0.28/M +local ~19GB 131k deepinfra.com ↗ 193 o1-preview 63±1 comparison only 194 Gemini 2.5 Flash Preview … 63±1 comparison only 195 GLM-4.5-Air 63±1 api $0.42/M +local ~35GB 131k z.ai ↗ 196 Qwen3.5 9B 63±1 api $0.11/M +local ~7.4GB 262k deepinfra.com ↗ 197 Seed-OSS 36B Instruct 63±1 local ~23GB 524k LM Studio › 198 GPT-4.5 Preview 63±1 comparison only 199 Trinity-Large-Thinking 63±2 api $0.39/M +local ~141GB 512k arcee.ai ↗ 200 Magistral Medium 1.2 (250… 63±2† comparison only 128k 201 MiniMax-M1 (80k) 62±1 api $0.96/M +local ~959GB 1M novita.ai ↗ 202 Gemma 4 12B 62±1 local ~8.5GB 262k LM Studio › 203 GPT-4.1 62±1 api $3.50/M 1M openai.com ↗ 204 Nemotron 3.5 Lightning 30… 62±2 api $0.087/M +loc… ~20GB 262k fireworks.ai ↗ 205 GLM-4.6V 62±2 api $0.45/M +local ~51GB 128k z.ai ↗ 206 Devstral 2 (2512) 62±2 local ~75GB 256k LM Studio › 207 Gemini 2.5 Flash Preview … 62±1 comparison only 208 A.X K1 (519B-A33B) 62±1† local ~1091GB how to run › 209 gpt-oss-120b 62±1 api $0.26/M +local ~67GB 131k groq.com ↗ 210 GPT-5 nano 62±1 api $0.14/M 400k openai.com ↗ 211 Qwen3-VL 32B Thinking 62±1 api $0.28/M +local ~22GB 256k alibabacloud.co… 212 Gemini 2.0 Flash-Lite (00… 62±2 comparison only 213 Mistral Large 3 (675B-A41… 62±1 api $0.75/M +local ~108GB 262k mistral.ai ↗ 214 Apriel 1.6 15B Thinker 62±2† local ~10GB LM Studio › 215 Gemini 2.5 Flash-Lite Pre… 62±2 comparison only 216 Gemini 2.5 Flash 62±1 api $0.85/M 1M google.dev ↗ 217 MiniMax-M1 (40k) 61±1† comparison only 218 Grok 4 Fast 61±2 comparison only 219 GLM-4.5V 61±2 api $0.90/M +local ~69GB 64k z.ai ↗ 220 Llama 3.1 Nemotron Ultra … 61±2 local ~82GB 128k llama.cpp › 221 HyperCLOVA X SEED 32B Thi… 61±2† comparison only 128k 222 Devstral Small 2 24B 61±3† local ~17GB 262k llama.cpp › 223 o1-mini 61±1 comparison only 224 Gemini 2.0 Pro Experiment… 61±1 comparison only 225 Qwen3 30B-A3B Thinking 25… 61±1 api $0.75/M +local ~20GB 262k alibabacloud.co… 226 Qwen3.5 4B 61±1 local ~4.3GB 262k llama.cpp › 227 Nemotron-Cascade-14B-Thin… 61±1† local ~10GB LM Studio › 228 Mellum2.1 12B-A2.5B Think… 60±2† local ~8.3GB LM Studio › 229 GPT-4.1 mini 60±1 api $0.70/M 1M openai.com ↗ 230 DeepSeek V3 0324 60±1 local ~108GB 164k llama.cpp › 231 Qwen3 30B-A3B Instruct 25… 60±1 api $0.15/M +local ~20GB 262k tokenfactory.ne… 232 Qwen3-VL 32B Instruct 60±1† api $0.28/M +local ~22GB 256k alibabacloud.co… 233 Falcon-H1R 7B 60±1† local ~6GB 262k LM Studio › 234 Gemini 2.0 Flash (001) 60±1 comparison only 235 ERNIE 4.5 300B-A47B 60±1† local ~97GB 131k llama.cpp › 236 Mistral Medium 3 (2505) 60±1 comparison only 237 Qwen3 235B-A22B (Apr 2025… 60±1 api $1.23/M +local ~76GB 131k alibabacloud.co… 238 Claude 3.5 Sonnet (2024-1… 60±1 comparison only 239 Granite 4.2 8B 60±1 api $0.11/M +local ~6.6GB 131k wandb.ai ↗ 240 Solar Open 100B (102B-A12… 60±1† local ~41GB 131k llama.cpp › 241 Qwen3-VL 30B-A3B Thinking 60±1† api $0.75/M +local ~21GB 256k alibabacloud.co… 242 Mistral Small 4 (119B-A6.… 60±2 api $0.26/M +local ~40GB 262k mistral.ai ↗ 243 Gemini 2.0 Flash-Lite Pre… 60±3 comparison only 244 QwQ-32B 59±1 local ~21GB 131k LM Studio › 245 OpenReasoning-Nemotron-32B 59±2† local ~21GB LM Studio › 246 DeepSeek V3 (Dec '24) 59±1 api $0.46/M +local ~337GB 131k deepinfra.com ↗ 247 Gemma 2 9B IT SimPO 59±1 local ~7.6GB llama.cpp › 248 Gemma 2 9B IT DPO (Prince… 59±1 local ~7.6GB llama.cpp › 249 Trinity-Large-Preview 59±2 local ~155GB llama.cpp › 250 GLM-4.7-Flash 59±1 api $0.0000/M +lo… ~19GB 200k z.ai ↗ 251 EXAONE 4.0 32B 59±1 local ~20GB 131k LM Studio › 252 Phi-4-reasoning-plus 14B 59±1 local ~10GB LM Studio › 253 Nemotron-Cascade-8B 59±1† local ~6.3GB LM Studio › 254 Qwen3 14B 59±1 api $0.61/M +local ~10GB 33k alibabacloud.co… 255 Mercury 2 59±2 comparison only 256 DiffusionGemma 26B-A4B 59±1† comparison only 262k 257 gpt-oss-20b 58±1 api $0.13/M +local ~14GB 131k groq.com ↗ 258 Llama 3.3 Nemotron Super … 58±2 local ~22GB llama.cpp › 259 Olmo 3.1 32B Think 58±1 local ~21GB 66k LM Studio › 260 DeepSeek R1 Distill Llama… 58±1 api $0.80/M +local ~30GB 131k novita.ai ↗ 261 Phi-4-reasoning 14B 58±1 api $0.22/M +local ~10GB 33k azure.microsoft… 262 Olmo 3 32B Think 58±1 local ~21GB 66k LM Studio › 263 Magistral Small 1.2 58±1 api $0.75/M +local ~16GB 131k aws.amazon.com ↗ 264 Nemotron-Cascade-8B-Think… 58±2† local ~6.3GB LM Studio › 265 Qwen3-VL 30B-A3B Instruct 58±1† api $0.35/M +local ~21GB 256k alibabacloud.co… 266 Qwen3-VL 8B Thinking 58±1 api $0.66/M +local ~6.8GB 256k alibabacloud.co… 267 Qwen3 32B 58±0 api $0.13/M +local ~20GB 131k deepinfra.com ↗ 268 Qwen3 Coder 30B-A3B Instr… 58±1 api $0.12/M +local ~20GB 160k novita.ai ↗ 269 Nova 2 Lite 58±2 api $0.94/M 1M aws.amazon.com ↗ 270 Llama 3.3 Nemotron Super … 58±1 local ~18GB 131k llama.cpp › 271 Nemotron 3 Nano 30B-A3B 58±1 api $0.087/M +loc… ~20GB 262k deepinfra.com ↗ 272 Ministral 3 14B Reasoning… 58±1 local ~10GB 262k llama.cpp › 273 GPT-4o (2024-08-06) 57±1 api $4.38/M 128k openai.com ↗ 274 Gemini 2.0 Flash Thinking… 57±2 comparison only 275 DeepSeek R1 0528 Qwen3 8B 57±1 api $0.068/M +loc… ~6.3GB 128k novita.ai ↗ 276 Granite 4.2 3B 57±1 api $0.052/M +loc… ~3.7GB 131k deepinfra.com ↗ 277 Command A (03-2025) 57±1 api $4.38/M +local ~37GB 256k azure.microsoft… 278 GPT-4o (2024-11-20) 57±1 api $4.38/M 128k openai.com ↗ 279 Gemini 2.5 Flash-Lite 57±2 api $0.17/M 1M google.dev ↗ 280 Qwen2.5-Max (2025-01-25) 57±1 comparison only 281 Qwen2.5-Omni 7B 57±2 local ~6.4GB llama.cpp › 282 GPT-4o (2024-05-13) 57±0 api $7.50/M 128k openai.com ↗ 283 QED-Nano 57±2 local ~4.1GB 262k llama.cpp › 284 Llama 4 Maverick (402B-A1… 57±1 api $0.42/M +local ~65GB 1M aws.amazon.com ↗ 285 Claude 3.5 Sonnet (2024-0… 57±1 comparison only 286 Qwen2.5 Coder 32B Instruct 57±1 api $0.74/M +local ~21GB 131k cloudflare.com ↗ 287 Qwen3 4B Thinking 2507 57±1 local ~3.9GB 262k LM Studio › 288 Gemini 1.5 Pro Experiment… 56±1 comparison only 289 Gemini 1.5 Pro (002) 56±1 comparison only 290 Athene-V2-Chat 72B 56±2 local ~44GB 33k LM Studio › 291 Gemini 2.0 Flash Experime… 56±1 comparison only 292 Sky-T1 32B Preview 56±5 local ~21GB LM Studio › 293 Qwen Plus (2025-01-25) 56±1 api $0.60/M 131k alibabacloud.co… 294 Grok 2 (1212) 56±1 comparison only 295 Qwen3-VL 4B Thinking 56±1 local ~4.1GB 256k llama.cpp › 296 EXAONE Deep 32B 56±2† local ~22GB 33k Ollama › 297 DeepSeek R1 Distill Qwen … 56±1 api $1.59/M +local ~21GB 131k cloudflare.com ↗ 298 Qwen2.5-VL 72B Instruct 56±1 local ~46GB 33k LM Studio › 299 Granite 4.1 30B 56±1† local ~19GB Ollama › 300 Mistral Small 3.2 24B 56±1 api $0.11/M +local ~16GB 128k deepinfra.com ↗ 301 DeepSeek-V2.5 (1210) 56±1 local ~82GB llama.cpp › 302 Qwen3 8B 56±1 api $0.31/M +local ~6.1GB 33k alibabacloud.co… 303 DeepSeek R1 Distill Qwen … 56±1 local ~10GB 131k LM Studio › 304 Ministral 3 14B 56±1 api $0.20/M +local ~9.1GB 262k mistral.ai ↗ 305 Nemotron Nano 9B v2 56±1 api $0.10/M +local ~7.6GB 131k aws.amazon.com ↗ 306 Magistral Medium 1.0 (250… 56±1 comparison only 307 GPT-4 Turbo (2024-04-09) 55±0 api $15.00/M 128k openai.com ↗ 308 Yi-Lightning 55±1 api price unknown 16k lingyiwanwu.com… 309 Qwen2.5 72B Instruct 55±1 api $0.37/M +local ~30GB 33k deepinfra.com ↗ 310 Mistral Large 2 (2411) 55±1 local ~78GB 128k llama.cpp › 311 Ministral 3 8B Reasoning … 55±1 local ~6.5GB 262k LM Studio › 312 Magistral Small 1.0 (2506) 55±1 local ~16GB 128k LM Studio › 313 Qwen3 30B-A3B (Apr 2025, … 55±1 api $0.35/M +local ~19GB 33k alibabacloud.co… 314 Molmo 2 8B 55±2 local ~6.8GB llama.cpp › 315 Magistral Medium 1.1 (250… 55±2† comparison only 40k 316 Gemini 1.5 Pro Experiment… 55±1 comparison only 317 Qwen3 4B Instruct 2507 55±1 local ~3.9GB 262k LM Studio › 318 Qwen2.5 Coder 14B Instruct 55±1 local ~10GB 131k LM Studio › 319 Qwen3-VL 8B Instruct 55±1 api $0.31/M +local ~6.8GB 256k alibabacloud.co… 320 Spark-X2.5 1.7B 55±1† local ~2.7GB 1M llama.cpp › 321 GigaChat 3.5 Ultra Instru… 55±1† local ~457GB 262k how to run › 322 Llama 3.1 405B Instruct 55±1 api $2.40/M +local ~160GB 128k aws.amazon.com ↗ 323 Qwen2.5 32B Instruct 55±1 local ~21GB 131k LM Studio › 324 Command A Reasoning (08-2… 55±2 local ~34GB llama.cpp › 325 Bielik 11B v3.0 Instruct 55±2 local ~8.1GB LM Studio › 326 Olmo 3.1 32B Instruct 55±2 local ~21GB 66k LM Studio › 327 OpenThinker 32B 55±1† local ~21GB 33k LM Studio › 328 Magistral Small 1.1 (2507) 55±2† local ~16GB 40k LM Studio › 329 EXAONE Deep 7.8B 55±2† local ~6.5GB 33k llama.cpp › 330 Qwen2.5-Math 72B Instruct 55±2† local ~44GB 4k LM Studio › 331 OpenThinker2 32B 54±1 local ~21GB 33k LM Studio › 332 Falcon-H1 34B Instruct 54±3 local ~23GB 262k llama.cpp › 333 Llama 3.3 70B Instruct 54±1 api $0.72/M +local ~24GB 128k aws.amazon.com ↗ 334 Llama 4 Scout (109B-A17B) 54±1 api $0.15/M +local ~36GB 328k deepinfra.com ↗ 335 Nemotron 3 Nano 4B 54±1 local ~3.8GB 262k LM Studio › 336 Granite 4.1 8B 54±1 api $0.11/M +local ~7GB 131k replicate.com ↗ 337 Pixtral Large (2411) 54±3 local ~51GB llama.cpp › 338 Bielik 11B v2.6 Instruct 54±3 local ~8.1GB LM Studio › 339 Nova Premier 54±2† api $5.00/M 1M aws.amazon.com ↗ 340 Mellum2 12B-A2.5B Thinking 54±1† local ~8.6GB 131k LM Studio › 341 Mistral Large 2 (2407) 54±1 local ~45GB 128k llama.cpp › 342 Mistral Small 3.1 24B (25… 54±1 local ~17GB 128k llama.cpp › 343 GPT-4 Turbo Preview (1106) 54±1 comparison only 344 GPT-4 Turbo Preview (0125) 54±1 comparison only 345 Claude 3.5 Haiku 54±1 comparison only 346 Llama 3.1 Tulu 3 70B 54±1 local ~43GB 131k LM Studio › 347 DeepSeek-V2.5 54±1 local ~151GB 164k Ollama › 348 DeepSeek-V2 Chat (0628) 54±1 local ~66GB llama.cpp › 349 GLM-4 9B Chat 1M 53±5 local ~8.1GB 1M llama.cpp › 350 Mistral Small 3 24B (2501) 53±1 local ~16GB 32k LM Studio › 351 Qwen2-VL 72B Instruct 53±1 local ~34GB llama.cpp › 352 Gemini 1.5 Flash Experime… 53±1 comparison only 353 GPT-4o mini (2024-07-18) 53±0 api $0.26/M 128k openai.com ↗ 354 Gemini 1.5 Pro (001) 53±1 comparison only 355 Llama 3.1 Tulu 3 70B DPO 53±1 local ~22GB 131k llama.cpp › 356 Qwen Turbo (2024-11-01) 53±2 comparison only 357 Qwen2.5-VL 32B Instruct 53±2 local ~22GB 33k LM Studio › 358 DeepSeek-Coder-V2 Instruc… 53±1 local ~92GB llama.cpp › 359 s1.1 32B 53±1 local ~22GB 33k LM Studio › 360 DeepSeek-Coder-V2 Instruc… 53±1 local ~151GB llama.cpp › 361 Olmo 3 7B Think 53±1 local ~5.8GB 66k LM Studio › 362 GPT-4.1 nano 53±1 api $0.17/M 1M openai.com ↗ 363 Gemini 1.5 Flash (002) 53±1 comparison only 364 Gemma 3 27B 53±1 api $0.10/M +local ~18GB 131k deepinfra.com ↗ 365 Phi-4 14B 53±1 api $0.087/M +loc… ~10GB 16k deepinfra.com ↗ 366 Gemma 4 E4B 53±1 local ~6.9GB 131k LM Studio › 367 Ministral 3 8B 53±1 api $0.15/M +local ~6.5GB 262k mistral.ai ↗ 368 Qwen3-VL 4B Instruct 53±1 local ~4.1GB 256k llama.cpp › 369 Claude 3 Opus 53±0 comparison only 370 LFM2.5 2.6B 53±1 local ~3.2GB 131k LM Studio › 371 Bielik 11B v2.1 Instruct 53±3 local ~8.5GB llama.cpp › 372 Llama 3.1 70B Instruct 52±1 api $0.72/M +local ~27GB 128k aws.amazon.com ↗ 373 Yi-Large Preview 52±1 comparison only 374 Qwen2.5-Math 7B Instruct 52±2† local ~6GB 4k LM Studio › 375 LFM2.5 8B-A1B 52±1 local ~6.5GB 128k LM Studio › 376 GPT-4 (0314) 52±1 comparison only 377 GLM-4 (0520) 52±1 comparison only 378 ERNIE 4.5 21B-A3B 52±1 comparison only 379 Bielik 11B v2.2 Instruct 52±3 local ~8.5GB llama.cpp › 380 Granite 4.0 H Small (32B-… 52±1 api $0.11/M +local ~21GB 131k replicate.com ↗ 381 Llama 3.1 Nemotron 70B In… 52±1 local ~43GB 128k LM Studio › 382 Palmyra X4 (palmyra-x-004) 52±4 comparison only 383 Nemotron-4 340B Instruct 52±2 comparison only 384 Aya Expanse 32B 52±2 api $0.75/M +local ~22GB 128k cohere.com ↗ 385 Gemma 3 12B 52±1 api $0.075/M +loc… ~9.1GB 131k deepinfra.com ↗ 386 Qwen2.5 Coder 7B Instruct 52±1 local ~6GB 131k LM Studio › 387 Qwen2.5 14B Instruct 52±1 local ~10GB 131k LM Studio › 388 Cogito v1 Preview Llama 8B 51±5 local ~6.2GB 128k LM Studio › 389 Hermes 3 Llama 3.1 405B 51±1 local ~114GB 131k llama.cpp › 390 GPT-4 (0613) 51±1 api $37.50/M 8k openai.com ↗ 391 OpenCoder-8B-Instruct 51±1 local ~6.4GB llama.cpp › 392 Command A Translate (08-2… 51±5† api price unknown 8k cohere.com ↗ 393 Llama 3.1 Nemotron 51B In… 51±2 local ~34GB llama.cpp › 394 Ministral 3 3B Reasoning … 51±1 local ~3.5GB 262k LM Studio › 395 Qwen3 4B 51±1 local ~3.7GB 33k LM Studio › 396 Yi-Large 51±1 comparison only 397 Llama-3-Instruct-8B-SimPO… 51±1 local ~6.6GB llama.cpp › 398 LFM2 24B-A2B 51±1 local ~16GB 33k LM Studio › 399 AceReason-Nemotron-1.0-14B 51±2† local ~10GB LM Studio › 400 Llama-3-Instruct-8B-SimPO 51±1 local ~6.6GB llama.cpp › 401 Athene-70B 51±1 local ~29GB llama.cpp › 402 Qwen2 72B Instruct 51±1 local ~33GB 131k llama.cpp › 403 Bielik 11B v2.0 Instruct 51±3 local ~8.5GB llama.cpp › 404 OpenThinker3 7B 50±1 local ~6GB 33k LM Studio › 405 CodeQwen1.5 7B Chat 50±1 local ~6.4GB 64k llama.cpp › 406 Jamba 1.5 Large 50±1 local ~839GB how to run › 407 Gemini 1.5 Flash (001) 50±1 comparison only 408 Llama-3-8B-ProLong-512k-I… 50±5 local ~35GB how to run › 409 EXAONE Deep 2.4B 50±2† local ~3.2GB 33k Ollama › 410 Mellum2 12B-A2.5B Instruct 50±2† local ~8.6GB 131k LM Studio › 411 OpenCodeInterpreter-DS-33B 50±1 local ~22GB LM Studio › 412 Llama 3.2 90B Vision Inst… 50±1 local ~58GB 128k Ollama › 413 DeepSeek-Coder 33B Instru… 50±1 local ~21GB 16k LM Studio › 414 Yi 34B 200K (base) 50±5 local ~23GB 200k llama.cpp › 415 Gemma 2 27B IT 50±1 local ~18GB 8k LM Studio › 416 Bielik 11B v2.3 Instruct 50±2 local ~8.6GB llama.cpp › 417 Qwen2.5-VL 7B Instruct 50±1 local ~6.4GB llama.cpp › 418 Molmo 72B (0924) 50±2 local ~309GB how to run › 419 Codestral 22B v0.1 50±1 local ~15GB 32k LM Studio › 420 GLM-4 (0116) 50±1 comparison only 421 Llama 3 70B Instruct 50±0 local ~24GB 8k llama.cpp › 422 WizardLM-2 8x22B 50±1 local ~91GB Ollama › 423 Nova Pro 50±1 api $1.40/M 300k aws.amazon.com ↗ 424 WizardCoder 33B V1.1 50±1 local ~22GB 16k llama.cpp › 425 Mistral Small Instruct 22… 49±2 local ~15GB 32k LM Studio › 426 Qwen2-Math 72B Instruct 49±1 local ~32GB 4k llama.cpp › 427 OLMo 2 32B Instruct (0325) 49±1 local ~21GB LM Studio › 428 OpenThinker2 7B 49±2† local ~6.4GB 33k llama.cpp › 429 Aya 23 35B 49±3 local ~22GB 8k LM Studio › 430 Granite 4.1 3B 49±1† local ~3.7GB Ollama › 431 OpenCodeInterpreter-DS-6.… 49±1 local ~5.8GB llama.cpp › 432 Artigenz-Coder-DS-6.7B 49±1 local ~5.8GB llama.cpp › 433 LFM2.5 1.2B Thinking 49±2 local ~2.2GB 33k LM Studio › 434 Qwen3-VL 2B Thinking 49±1 local ~2.7GB 256k llama.cpp › 435 speechless-codellama-34b-… 49±1 local ~23GB llama.cpp › 436 Pixtral 12B (2409) 49±2 local ~9.6GB llama.cpp › 437 Nova Lite 49±2 api $0.10/M 300k aws.amazon.com ↗ 438 MegaBeam-Mistral-7B-512k 49±4 local ~6.1GB llama.cpp › 439 GPT-4o Audio Preview (202… 49±2 comparison only 440 Claude 3 Sonnet 49±1 comparison only 441 AceReason-Nemotron-1.1-7B 49±2† local ~6GB LM Studio › 442 Mixtral 8x22B Instruct v0… 49±1 local ~56GB 64k llama.cpp › 443 Yi-Coder 9B Chat 49±1 local ~6.8GB 128k LM Studio › 444 Qwen2-Math 7B Instruct 49±2† local ~6GB 4k LM Studio › 445 Magicoder-S-DS 6.7B 49±1 local ~5.8GB llama.cpp › 446 InternLM2.5 7B Chat 49±2 local ~6.1GB LM Studio › 447 Hermes 3 Llama 3.1 70B 49±1 local ~44GB 131k Ollama › 448 Qwen2.5 7B Instruct 48±1 api $0.30/M +local ~6GB 131k together.ai ↗ 449 DeepSeek R1 Distill Qwen … 48±1 local ~6GB 131k LM Studio › 450 XwinCoder-34B 48±1 local ~23GB llama.cpp › 451 StarChat2 15B v0.1 48±1 local ~12GB 16k llama.cpp › 452 Code Millenials 34B 48±1 local ~23GB llama.cpp › 453 Reka Core (2024-05-01) 48±1 comparison only 454 Gemini 1.5 Flash-8B (001) 48±2 comparison only 455 T5 Large 48±9 local ~4.6GB 1k how to run › 456 WaveCoder-Ultra-6.7B 48±1 local ~5.8GB llama.cpp › 457 Phi-3.5-MoE-instruct 48±1 local ~26GB 128k LM Studio › 458 Claude 3 Haiku 48±0 comparison only 459 GPT-4o mini Audio Preview… 48±2 comparison only 460 Gemma 3n E4B 48±1 local ~6.2GB 32k llama.cpp › 461 Hermes 2 Theta Llama-3 70B 48±2 local ~47GB 8k llama.cpp › 462 Granite 4.0 H Tiny (7B-A1… 48±1 local ~5.6GB 131k LM Studio › 463 DeepSeek-Coder 7B Instruc… 48±1 local ~5.9GB llama.cpp › 464 Qwen2 57B-A14B Instruct 48±2 local ~35GB 33k LM Studio › 465 Qwen2.5 Coder 3B Instruct 48±1 local ~3.3GB 33k LM Studio › 466 DeepSeek-Coder 6.7B Instr… 48±1 local ~5.8GB 16k llama.cpp › 467 MiniCPM3-4B 48±2 local ~4GB 32k LM Studio › 468 Bielik 4.5B v3.0 Instruct 48±2 local ~4.3GB LM Studio › 469 Qwen3.5 2B 47±1 local ~2.9GB 262k llama.cpp › 470 Gemma 2 9B IT 47±1 local ~7GB 8k LM Studio › 471 Command R Plus (08-2024) 47±1 api $4.38/M +local ~63GB 128k cohere.com ↗ 472 GLM-4 9B Chat 47±2 local ~7.9GB 128k Ollama › 473 Yi-1.5 34B (base) 47±3 local ~22GB 4k LM Studio › 474 Phi-4-mini-reasoning 3.8B 47±2 api $0.13/M +local ~3.8GB 128k azure.microsoft… 475 Qwen1.5 110B Chat 47±1 local ~45GB 32k llama.cpp › 476 WhiteRabbitNeo-33B-v1 47±1 local ~22GB llama.cpp › 477 DeepSeek R1 Distill Llama… 47±1 local ~6.2GB 131k LM Studio › 478 EXAONE 4.0 1.2B 47±2† local ~2.2GB 66k LM Studio › 479 InternVL2-26B 47±2 local ~55GB how to run › 480 speechless-coder-ds-6.7b 47±1 local ~5.8GB llama.cpp › 481 Gemma 4 E2B 47±1 api $0.050/M +loc… ~5.1GB 131k aws.amazon.com ↗ 482 Magicoder-S-CL-7B 47±1 local ~5.8GB llama.cpp › 483 Aya Expanse 8B 47±2 local ~6.2GB 8k LM Studio › 484 Phind CodeLlama 34B v2 47±1 local ~23GB 16k Ollama › 485 CogVLM2 Llama3 Chinese Ch… 47±7† comparison only 486 InternLM2.5 7B Chat 1M 47±4 local ~6.4GB llama.cpp › 487 EuroLLM 9B Instruct 47±3 local ~6.9GB 4k LM Studio › 488 Granite 4.0 H Micro 3B 47±1 local ~3.4GB 131k LM Studio › 489 GPT-3.5 Turbo (0125) 46±1 api $0.75/M 16k openai.com ↗ 490 Mistral Large (2402) 46±1 comparison only 491 Jamba 1.5 Mini 46±1 local ~34GB llama.cpp › 492 MedGemma 4B IT 46±3 local ~4.1GB llama.cpp › 493 OLMo 2 13B Instruct (1124) 46±1 local ~9.6GB 4k LM Studio › 494 Falcon-H1 7B Instruct 46±3 local ~6GB 262k LM Studio › 495 GPT-3.5 Turbo (1106) 46±1 comparison only 496 Cogito v1 Preview Llama 3B 46±4 local ~3.4GB 128k LM Studio › 497 Code Llama 70B Instruct 46±1 local ~28GB 16k llama.cpp › 498 Hermes 2 Pro Llama 3 8B 46±3 local ~6.6GB 8k llama.cpp › 499 Llama 3 70B (base) 46±2 local ~43GB LM Studio › 500 Granite 4.0 Micro 3B 46±1 local ~3.5GB 131k LM Studio › 501 Command R (08-2024) 46±1 api $0.26/M +local ~21GB 128k cohere.com ↗ 502 GPT-3.5 Turbo (0301) 46±1 comparison only 503 Nova Micro 46±2 api $0.061/M 128k aws.amazon.com ↗ 504 T5 Base 46±9 local ~2.4GB 1k how to run › 505 OpenChat 3.5 0106 46±1 local ~6.1GB 8k llama.cpp › 506 Qwen2-VL 7B Instruct 46±1 local ~6.4GB LM Studio › 507 Qwen2.5-Math 1.5B Instruct 46±2† local ~2.4GB 4k LM Studio › 508 Yi 1.5 34B Chat 46±1 local ~22GB 4k LM Studio › 509 DBRX Instruct 46±1 local ~85GB 32k Ollama › 510 Granite 34B Code Instruct… 46±1 local ~22GB Ollama › 511 WizardCoder Python 34B V1… 46±1 local ~23GB 16k Ollama › 512 Gemini 1.0 Pro (001) 46±2 comparison only 513 Mistral Medium (2312) 46±1 comparison only 514 GPT-3.5 Turbo (0613) 46±1 comparison only 515 Granite 3.3 8B Instruct 46±1 local ~6.3GB 128k LM Studio › 516 Tulu 2 70B 46±2† local ~28GB llama.cpp › 517 LFM2 8B-A1B 46±1 comparison only 518 OpenChat 3.5 0106 Gemma 46±3 local ~6.9GB 8k llama.cpp › 519 Olmo 3 7B Instruct 46±1 local ~5.8GB 66k LM Studio › 520 InternLM2.5 20B Chat 46±2 local ~14GB 33k llama.cpp › 521 StarCoder2-15B-Instruct-v… 46±1 local ~11GB 16k Ollama › 522 Qwen1.5 72B Chat 45±1 local ~31GB 32k llama.cpp › 523 speechless-starcoder2-15b 45±1 local ~12GB llama.cpp › 524 Ministral 3 3B 45±1 api $0.10/M +local ~3.5GB 262k mistral.ai ↗ 525 Claude 2.0 45±1 comparison only 526 OpenThinker 7B 45±2† local ~6GB 33k LM Studio › 527 Phi-3 Medium 128K Instruct 45±1 local ~9.8GB 128k LM Studio › 528 QwQ-32B-Preview 45±2 local ~21GB 33k LM Studio › 529 DeepSeek-Coder-V2 Lite In… 45±1 local ~11GB 128k LM Studio › 530 GPT-4o Audio Preview (202… 45±3 comparison only 531 Claude 2.1 45±1 comparison only 532 Zephyr ORPO 141B-A35B v0.1 45±2 local ~56GB 66k llama.cpp › 533 Phi-3-medium-4k-instruct 45±1 local ~9.8GB 4k LM Studio › 534 Qwen2-Math 1.5B Instruct 45±2† local ~2.4GB 4k LM Studio › 535 Mistral Small (2402) 45±1 comparison only 536 Llama 3.1 8B Instruct 45±0 api $0.22/M +local ~6.2GB 128k aws.amazon.com ↗ 537 Command R Plus (04-2024) 45±1 local ~43GB 128k llama.cpp › 538 Mistral NeMo Instruct (24… 44±1 local ~8.7GB 128k LM Studio › 539 LLaVA-OneVision Qwen2 72B… 44±3 local ~155GB how to run › 540 Phi-3-mini-4k-instruct (J… 44±1 local ~3.7GB LM Studio › 541 Yi 1.5 9B Chat 44±1 local ~6.8GB 4k LM Studio › 542 Command R7B (12-2024) 44±1 local ~6.2GB 128k LM Studio › 543 Qwen2-Audio 7B Instruct 44±3 local ~6.5GB llama.cpp › 544 Gemma 3 4B 44±1 api $0.063/M +loc… ~3.8GB 131k deepinfra.com ↗ 545 Eurus-2-7B-PRIME 44±2 local ~6.4GB 4k llama.cpp › 546 Llama 3.1 Tulu 3 8B 44±1 local ~6.2GB 131k LM Studio › 547 Qwen2.5-VL 3B Instruct 44±1 local ~3.5GB llama.cpp › 548 Qwen1.5 32B Chat 44±1 local ~22GB 32k llama.cpp › 549 Granite 20B Code Instruct… 44±1 local ~14GB Ollama › 550 Qwen2 7B Instruct 44±1 local ~6GB 131k LM Studio › 551 Falcon3 7B Instruct 44±3 local ~5.9GB 32k LM Studio › 552 Xwin-LM 70B v0.1 44±1 local ~28GB llama.cpp › 553 Qwen3-VL 2B Instruct 44±1 local ~2.7GB 256k llama.cpp › 554 Dolphin 2.6 Mixtral 8x7B 44±1 local ~29GB llama.cpp › 555 Granite 8B Code Instruct … 44±1 local ~6.6GB llama.cpp › 556 Falcon3 10B Instruct 44±3 local ~7.6GB 32k LM Studio › 557 Code Tulu 2 34B 44±2† local ~72GB how to run › 558 Qwen3 1.7B 44±1 local ~2.6GB 33k LM Studio › 559 Phi-4-mini 3.8B 44±1 api $0.13/M +local ~3.8GB 128k azure.microsoft… 560 Ministral 8B (2410) 44±1 local ~6.2GB 128k LM Studio › 561 Granite 3.2 8B Instruct 44±1 local ~6.6GB 131k llama.cpp › 562 Code Llama 70B Python 44±1 local ~28GB 16k llama.cpp › 563 Llama 3.1 8B (base) 44±3 local ~6.2GB LM Studio › 564 LFM2.5 1.2B Instruct 44±2 local ~2.2GB 33k LM Studio › 565 Granite 3.0 8B Instruct 43±1 local ~6.6GB 4k llama.cpp › 566 Phi-3.5-mini-instruct 43±1 local ~3.7GB 128k LM Studio › 567 Hermes 2 Theta Llama 3 8B 43±2 local ~6.6GB 8k llama.cpp › 568 Yi 34B Chat 43±1 local ~23GB 4k llama.cpp › 569 CogVLM2 Llama3 Chat 19B 43±3 local ~43GB how to run › 570 Code Llama 70B (base) 43±1 local ~28GB 16k llama.cpp › 571 Gemma 3n E2B IT 43±2 local ~4.7GB 32k llama.cpp › 572 Reka Flash 21B (2024-02-2… 43±1 comparison only 573 T5 Small 43±9 local ~1.8GB 1k how to run › 574 MiniCPM-V 2.6 43±2 local ~6.4GB llama.cpp › 575 Qwen2.5 Coder 1.5B Instru… 43±1 local ~2.4GB 33k LM Studio › 576 LFM2 2.6B 43±2 local ~3GB 33k LM Studio › 577 SmolLM3 3B 43±1 local ~3.3GB 66k LM Studio › 578 Granite 3.1 8B Instruct 43±1 local ~6.6GB 128k llama.cpp › 579 Llama 3.2 11B Vision Inst… 43±2 local ~8.8GB 128k Ollama › 580 Qwen2.5 3B Instruct 43±1 local ~3.3GB 33k LM Studio › 581 DeepSeek-Coder 1.3B Instr… 43±1 local ~2.4GB 16k Ollama › 582 Aya 23 8B 43±3 local ~6.2GB 8k LM Studio › 583 Phi-3-small-8k-instruct 43±2 local ~17GB how to run › 584 CodeGemma 7B IT (v1.0) 43±1 local ~7.1GB llama.cpp › 585 Aya Vision 32B 42±3 api price unknown… ~71GB 16k cohere.com ↗ 586 Tulu 2 DPO 70B 42±1 local ~32GB llama.cpp › 587 Mathstral 7B v0.1 42±1 local ~5.8GB 32k LM Studio › 588 Yi-1.5 9B (base) 42±2 local ~6.8GB 4k LM Studio › 589 Claude Instant 1.2 42±1 comparison only 590 Llama 3 8B Instruct 42±0 local ~6.6GB 8k llama.cpp › 591 WizardLM 70B V1.0 42±1 local ~32GB llama.cpp › 592 WizardCoder Python 13B V1… 42±2 local ~9.8GB 16k llama.cpp › 593 Hermes 3 Llama 3.1 8B 42±1 local ~6.6GB 128k llama.cpp › 594 Falcon2 11B (base) 42±3 local ~8.2GB 8k Ollama › 595 Granite 34B Code Base (8K) 42±1 local ~24GB llama.cpp › 596 speechless-starcoder2-7b 42±1 local ~6.1GB llama.cpp › 597 Command R (03-2024) 42±1 local ~24GB 128k llama.cpp › 598 Ministral 3B (2410) 42±1 comparison only 599 OLMo 2 7B Instruct (1124) 42±1 local ~5.8GB 4k LM Studio › 600 WizardCoder-15B-V1.0 42±1 local ~12GB llama.cpp › 601 Snowflake Arctic Instruct 41±2 local ~1007GB how to run › 602 Phi-3-vision-128k-instruct 41±2 local ~10GB how to run › 603 OpenChat 3.5 41±2 local ~6.1GB 8k llama.cpp › 604 InternVL2-4B 41±2 local ~10GB how to run › 605 Qwen1.5 14B Chat 41±1 local ~11GB 32k Ollama › 606 Code Llama 34B Python 41±1 local ~23GB 16k Ollama › 607 speechless-coding-7b-16k-… 41±1 local ~16GB how to run › 608 DeepSeek-Coder 33B Base 41±1 local ~22GB 16k llama.cpp › 609 InternLM2 Chat 20B 41±1 local ~14GB llama.cpp › 610 DeepSeek LLM 67B Chat 41±1 local ~31GB 4k llama.cpp › 611 Llama 2 70B (base) 41±1 local ~32GB 4k llama.cpp › 612 Mixtral 8x7B v0.1 (base) 41±1 local ~29GB 32k llama.cpp › 613 North Micro Vision Instru… 41±2† local ~3.8GB 128k how to run › 614 CodeQwen1.5 7B (base) 41±1 local ~6.4GB 66k llama.cpp › 615 Granite 4.0 1B 41±1 local ~2.4GB 128k LM Studio › 616 Mixtral 8x7B Instruct v0.1 41±1 local ~29GB 32k LM Studio › 617 Molmo 7B-D (0924) 41±1 local ~35GB how to run › 618 OpenHermes-2.5-Code-290k-… 41±1 local ~9.8GB llama.cpp › 619 Granite 4.0 H 1B 40±1 local ~2.3GB 131k LM Studio › 620 Llama3-OpenBioLLM-8B 40±2 local ~6.6GB llama.cpp › 621 Nous Hermes 2 Mistral 7B … 40±2 local ~6.1GB 33k llama.cpp › 622 Nous Hermes 2 Mixtral 8x7… 40±2 local ~31GB 33k llama.cpp › 623 Deita 7B v1.0 40±2 local ~32GB how to run › 624 Code Llama 34B Instruct 40±1 local ~22GB 16k LM Studio › 625 Code Llama 34B (base) 40±1 local ~23GB 16k Ollama › 626 Starling-LM 7B beta 40±1 local ~6GB 8k LM Studio › 627 LFM2 1.2B 40±2 local ~2.2GB 33k LM Studio › 628 Code Tulu 2 13B 40±2† local ~29GB how to run › 629 LLaVA-v1.6-34B 40±2 local ~24GB 4k Ollama › 630 Bielik 1.5B v3.0 Instruct 40±3 local ~2.4GB LM Studio › 631 OpenHermes 2.5 Mistral 7B 39±1 local ~6.1GB llama.cpp › 632 Granite 3.3 2B Instruct 39±1† local ~3GB 128k LM Studio › 633 Phi-3-mini-128k-instruct … 39±2 local ~9.5GB how to run › 634 Mistral 7B Instruct v0.3 39±1 local ~5.8GB 32k LM Studio › 635 Vicuna 33B v1.3 39±1 local ~23GB 2k Ollama › 636 DeepSeek-Coder 6.7B Base 39±1 local ~5.8GB 16k llama.cpp › 637 Starling-LM 7B alpha 39±1 local ~6.1GB 8k llama.cpp › 638 Granite 8B Code Base (4K) 39±1 local ~6.6GB llama.cpp › 639 StarCoder2 15B 39±1 local ~11GB 16k LM Studio › 640 speechless-code-mistral-7… 39±1 local ~17GB how to run › 641 Falcon-H1 3B Instruct 39±4 local ~3.4GB 131k LM Studio › 642 Phi-2 (2.7B) 39±1 local ~3.2GB 2k LM Studio › 643 Granite 3.0 2B Instruct 39±1 local ~3.2GB 4k llama.cpp › 644 Llama 2 70B Chat 39±1 local ~32GB 4k llama.cpp › 645 Code Llama 13B Instruct 39±1 local ~9.8GB 16k llama.cpp › 646 MistralHermes-CodePro-7B-… 39±1 local ~6.1GB llama.cpp › 647 Llama-3-Instruct-8B-SimPO… 39±1 local ~6.6GB llama.cpp › 648 Llama 3.2 3B Instruct 38±1 api $0.12/M +local ~3.4GB 131k cloudflare.com ↗ 649 Qwen-14B-Chat 38±1 local ~11GB llama.cpp › 650 Falcon-180B-Chat 38±2 local ~115GB 2k Ollama › 651 CodeGemma 7B (v1.0) 38±1 local ~7.1GB llama.cpp › 652 Code Llama 13B Python 38±1 local ~9.8GB 16k llama.cpp › 653 Tulu 2 13B 38±2 local ~9.8GB llama.cpp › 654 Phi-3-mini-4k-instruct (A… 38±2 local ~4GB llama.cpp › 655 Apertus 8B Instruct (2509) 38±4 local ~6.3GB 66k LM Studio › 656 LLaMA 65B 38±1 comparison only 657 Granite 3B Code Instruct … 38±2 local ~3.7GB llama.cpp › 658 DeepSeek-V2 Lite Chat (16… 38±1 local ~11GB 32k LM Studio › 659 BioMistral-7B 38±3 local ~6.1GB llama.cpp › 660 Qwen3.5 0.8B 38±1 local ~2GB 262k llama.cpp › 661 Code Tulu 2 7B 38±2† local ~16GB how to run › 662 SOLAR 10.7B Instruct v1.0 37±1 local ~8.3GB 4k llama.cpp › 663 Phi-3.5-vision-instruct 37±2 local ~10GB how to run › 664 Tulu 2 DPO 13B 37±1 local ~9.8GB llama.cpp › 665 Yi-1.5 6B (base) 37±3 local ~5.1GB 4k LM Studio › 666 Qwen2-VL 2B Instruct 37±1 local ~2.8GB 33k LM Studio › 667 DeepSeek R1 Distill Qwen … 37±2 local ~2.6GB 131k LM Studio › 668 Gemma 2 2B IT 37±1 local ~3.1GB 8k LM Studio › 669 Qwen2.5 Coder 0.5B Instru… 37±1 local ~1.8GB 33k LM Studio › 670 Granite 20B Code Base (8K) 37±1 local ~15GB llama.cpp › 671 Gemma 1.1 7B IT 37±1 local ~7.1GB 8k llama.cpp › 672 Code Llama 7B Python 37±1 local ~5.8GB 16k llama.cpp › 673 Dolphin 2.2.1 Mistral 7B 37±2 local ~6.1GB llama.cpp › 674 Code Llama 13B (base) 37±1 local ~9.2GB 16k LM Studio › 675 Qwen1.5 7B Chat 37±1 local ~6.5GB 32k llama.cpp › 676 Yi 1.5 6B Chat 37±2 local ~5.1GB 4k LM Studio › 677 Falcon3 3B Instruct 37±3 local ~3.4GB 32k LM Studio › 678 Zephyr 7B alpha 36±1 local ~6.1GB llama.cpp › 679 Vicuna 13B v1.5 36±2 local ~9.8GB 4k llama.cpp › 680 Granite 3.1 2B Instruct 36±1 local ~3.1GB 128k llama.cpp › 681 EXAONE 3.5 2.4B Instruct 36±2 local ~3GB 33k LM Studio › 682 LLaMA 33B 36±1 comparison only 683 Code Llama 7B Instruct 36±1 local ~5.5GB 16k LM Studio › 684 LFM2 700M 36±2 local ~1.9GB 33k LM Studio › 685 Granite 3.2 2B Instruct 36±1 local ~3GB 131k LM Studio › 686 Mistral 7B Instruct v0.2 36±1 local ~6GB 32k LM Studio › 687 WizardLM 13B V1.2 36±1 local ~9.8GB llama.cpp › 688 Tulu 2 7B 36±2 local ~5.8GB llama.cpp › 689 xDAN-L1-Chat-RL-v1 36±1 local ~6.1GB llama.cpp › 690 Llama 3 8B (base) 36±1 local ~6.6GB llama.cpp › 691 Mistral 7B v0.1 (base) 36±0 local ~5.8GB 8k LM Studio › 692 Qwen2.5 1.5B Instruct 36±1 local ~2.4GB 33k LM Studio › 693 MPT-30B-Chat 35±2 comparison only 694 Gemma 7B (base) 35±1 local ~7.1GB 8k llama.cpp › 695 Tulu 2 DPO 7B 35±1 local ~5.8GB llama.cpp › 696 Eurus-2-7B-SFT 35±3† local ~6.4GB llama.cpp › 697 Falcon 40B (base) 35±1 local ~27GB 2k Ollama › 698 StarCoder2 7B 35±1 local ~6.1GB 16k LM Studio › 699 Granite 3.0 3B-A800M Inst… 35±1 local ~3.7GB 4k llama.cpp › 700 DeepSeek LLM 7B Chat 35±2† local ~5.7GB Ollama › 701 StarCoder 15.5B 35±1 local ~11GB Ollama › 702 Granite 3.1 3B-A800M Inst… 35±2† local ~3.6GB 131k llama.cpp › 703 Llama 2 13B (base) 34±1 local ~9.3GB 4k Ollama › 704 Code Llama 7B (base) 34±1 local ~5.5GB 16k LM Studio › 705 Nous-Hermes-13b 34±2 local ~9.8GB 2k llama.cpp › 706 ChatGLM3 6B 128K 34±4 local ~5.8GB 131k llama.cpp › 707 Zephyr 7B beta 34±1 local ~6.1GB llama.cpp › 708 PaLM 2 Chat (chat-bison@0… 34±2 comparison only 709 Qwen3 0.6B 33±1 local ~1.8GB 33k LM Studio › 710 Llama 2 13B Chat 33±1 local ~9.8GB 4k llama.cpp › 711 Flan-T5 XXL 33±1 local ~49GB how to run › 712 StarCoder2 3B 33±1 local ~3.4GB 16k llama.cpp › 713 Granite 3B Code Base (2K) 33±1 local ~3.7GB llama.cpp › 714 GPT-3 davinci (API base m… 33±1 comparison only 715 StarCoderBase 15.5B 33±1 local ~12GB llama.cpp › 716 Vicuna 7B v1.5 33±1 local ~5.8GB 4k llama.cpp › 717 Guanaco 33B 33±2 local ~70GB how to run › 718 Stable Code 3B 33±1 local ~3.3GB 16k llama.cpp › 719 OLMo 2 1B Instruct (0425) 33±2 local ~2.4GB 4k llama.cpp › 720 DeepSeek-Coder 1.3B Base 32±1 local ~2.4GB 16k Ollama › 721 Orca 2 13B 32±2 local ~9.8GB 4k llama.cpp › 722 Gemma 7B IT 32±1 local ~7.1GB 8k llama.cpp › 723 Mistral 7B Instruct v0.1 32±1 local ~6.1GB 8k llama.cpp › 724 Qwen1.5 4B Chat 32±2 local ~4.1GB 32k llama.cpp › 725 Ghost 7B Alpha 32±2 local ~6.1GB llama.cpp › 726 CodeGemma 2B (v1.0) 32±1 local ~3.2GB llama.cpp › 727 Qwen-7B (base) 32±1 local ~6.6GB 33k llama.cpp › 728 Phi-1.5 (1.3B) 32±1 local ~2.4GB 2k llama.cpp › 729 Falcon-H1 1.5B Instruct 31±4 local ~2.4GB 131k LM Studio › 730 Flan-T5 XL 31±2 local ~14GB how to run › 731 Qwen2 1.5B Instruct 31±1 local ~2.4GB 32k LM Studio › 732 Qwen-7B-Chat 31±1 local ~6.6GB 33k llama.cpp › 733 Gemma 3 1B 31±1 local ~2.2GB 32k LM Studio › 734 StarCoderBase 7B 31±1 local ~6GB Ollama › 735 Llama 2 7B (base) 31±1 local ~5.5GB 4k Ollama › 736 OPT-66B 31±1 local ~141GB 2k how to run › 737 LFM2 350M 30±2 local ~1.7GB 33k LM Studio › 738 SmolLM2 1.7B Instruct 30±1 local ~2.6GB 8k LM Studio › 739 Llama 3.2 1B Instruct 30±1 api $0.020/M +loc… ~2.2GB 131k novita.ai ↗ 740 Falcon 7B (base) 30±1 local ~5.9GB 2k Ollama › 741 Llama 2 7B Chat 30±1 local ~5.8GB 4k llama.cpp › 742 Baichuan2-7B-Base 30±1 local ~17GB how to run › 743 LLaMA 7B 30±1 comparison only 744 MPT-7B (base) 30±1 comparison only 745 OLMo 7B Instruct 29±2 local ~16GB how to run › 746 StableLM-3B-4E1T 29±1 local ~3.3GB 4k llama.cpp › 747 Gemma 2B (base) 29±1 local ~3.2GB 8k llama.cpp › 748 OpenLLaMA 7B v2 29±1 local ~9.1GB llama.cpp › 749 Gemma 1.1 2B IT 29±1 local ~3.2GB 8k llama.cpp › 750 StableLM-Base-Alpha 7B v2 29±1 local ~30GB 4k how to run › 751 Granite 3.1 1B-A400M Inst… 29±2† local ~2.3GB 131k llama.cpp › 752 OPT-30B 29±1 local ~64GB 2k how to run › 753 NexusRaven V2 13B 29±3 local ~9.3GB 16k Ollama › 754 OLMo 7B 28±1† local ~30GB 2k how to run › 755 Falcon 40B Instruct 28±2 local ~27GB Ollama › 756 GPT-NeoX-20B 28±1 local ~14GB 2k LM Studio › 757 LLaMA 13B 28±1 comparison only 758 Koala 13B 28±1 comparison only 759 StarCoderBase 3B 28±1 local ~3.4GB Ollama › 760 Granite 3.0 1B-A400M Inst… 27±1 local ~2.3GB 4k llama.cpp › 761 RedPajama-INCITE 7B Base 27±1 local ~16GB 2k how to run › 762 ChatGLM3-6B 27±2 local ~5.8GB llama.cpp › 763 Granite 4.0 H 350M 27±1† local ~1.7GB 33k LM Studio › 764 Flan-T5 Large 27±2 local ~4.8GB how to run › 765 MPT-7B-Chat 27±2 comparison only 766 Gemma 2B IT 27±1 local ~3.1GB 8k llama.cpp › 767 EuroLLM 1.7B Instruct 27±3 local ~2.4GB 4k LM Studio › 768 Qwen2.5 0.5B Instruct 27±2 local ~1.8GB 33k LM Studio › 769 GPT-3 text-davinci-001 26±3 comparison only 770 GPT4All-13B-Snoozy 26±2 local ~56GB how to run › 771 OPT-13B (13 billion param… 26±1 local ~29GB 2k how to run › 772 Granite 4.0 350M 26±1 local ~1.7GB 32k LM Studio › 773 GPT-3 curie (API base mod… 26±1 comparison only 774 Alpaca 13B 26±2 comparison only 775 Falcon 7B Instruct 25±2 local ~5.9GB Ollama › 776 Qwen2 0.5B Instruct 25±2 local ~1.8GB 32k LM Studio › 777 Pythia 12B 25±1 local ~9.5GB 2k llama.cpp › 778 RedPajama-INCITE Base 3B … 25±3 local ~3.4GB 2k llama.cpp › 779 ChatGLM2-6B 25±2 local ~15GB how to run › 780 GPT-J 6B 25±1 local ~14GB 2k how to run › 781 OpenLLaMA 3B v2 25±1 local ~4.2GB llama.cpp › 782 OPT-6.7B 24±1 local ~16GB 2k how to run › 783 RWKV-4-Raven 14B 24±2 comparison only 784 Falcon3 1B Instruct 24±4 local ~2.4GB 8k LM Studio › 785 StableLM-Base-Alpha 3B v2 24±1 local ~13GB 4k how to run › 786 Qwen1.5 1.8B Chat 24±2 local ~2.7GB 33k Ollama › 787 Pythia 6.9B 23±1 local ~6.1GB 2k llama.cpp › 788 OpenAssistant SFT-4 Pythi… 22±1 local ~9.5GB llama.cpp › 789 Pythia 2.8B 21±1 local ~3.4GB 2k llama.cpp › 790 BLOOM 7.1B 21±1 local ~7.1GB 2k llama.cpp › 791 OLMo 1B 21±2† local ~2.2GB 2k llama.cpp › 792 Phi-1 (1.3B) 21±4 local ~2.4GB 2k llama.cpp › 793 Qwen1.5 0.5B Chat 21±3 local ~1.8GB 33k LM Studio › 794 ChatGLM-6B 20±2 local ~16GB how to run › 795 Pythia 2.8B (deduped) 20±1 local ~3.4GB 2k llama.cpp › 796 Flan-T5 Base 20±3† local ~2.5GB how to run › 797 Falcon-RW 1B 20±2 local ~4.3GB 2k how to run › 798 OPT-2.7B 19±1 local ~7.1GB 2k how to run › 799 SmolLM2 360M Instruct 19±3 local ~1.7GB 8k LM Studio › 800 GPT-Neo 2.7B 18±1 local ~13GB 2k how to run › 801 GPT-3 babbage (API base m… 18±1 comparison only 802 Pythia 1.4B 18±2 local ~2.4GB 2k llama.cpp › 803 BLOOM 3B 18±1 local ~4GB 2k llama.cpp › 804 FastChat-T5 3B 17±2 local ~8.5GB how to run › 805 Dolly v2 12B 17±2 comparison only 806 OPT-1.3B 16±1 local ~4.3GB 2k how to run › 807 Gemma 3 270M IT 15±4 local ~1.7GB 32k LM Studio › 808 Pythia 1B 14±1 local ~2.2GB 2k llama.cpp › 809 StableLM-Tuned-Alpha 7B 14±2 local ~6.9GB llama.cpp › 810 BLOOM 1.7B 14±1 local ~3.1GB 2k llama.cpp › 811 Flan-T5 Small 14±3† local ~1.8GB how to run › 812 GPT-Neo 1.3B 14±2 local ~7.1GB 2k how to run › 813 GPT-2 XL (1.5B) 11±1 local ~2.8GB 1k llama.cpp › 814 BLOOM 1.1B 11±1 local ~2.6GB 2k llama.cpp › 815 Pythia 410M 10±2 local ~1.8GB 2k llama.cpp › 816 GPT-3 ada (API base model) 10±2 comparison only 817 StableLM-Base-Alpha 7B 9±1 local ~35GB 4k how to run › 818 Zyte-1B 8±2 local ~2.2GB llama.cpp › 819 GPT-Neo 125M 5±5 local ~2.1GB 2k how to run › 820 BLOOM 560M 5±2 local ~2.1GB 2k llama.cpp › 821 OPT-350M 5±2 local ~2.2GB 2k how to run › 822 StableLM-Base-Alpha 3B 4±1 local ~17GB 4k how to run › 823 Pythia 160M 3±2 local ~1.6GB 2k llama.cpp › 824 OPT-125M 3±2 local ~1.8GB 2k how to run › 825 GPT-2 Large (774M) 1±2 local ~2GB 1k llama.cpp › 826 GPT-2 Medium (355M) -1±2 local ~1.8GB 1k llama.cpp › 827 Pythia 70M -7±2 local ~1.6GB 2k llama.cpp › 828 GPT-2 Small (124M) -9±3 local ~1.6GB 1k LM Studio › — Grok Build 0.1 no reviewed results api $1.25/M 256k x.ai ↗ — K2 Horizon 7B not scored yet local ~7.7GB 524k llama.cpp ›
+ 304 models with fewer than 3 results aren't ranked here · show them
* new or thinly tested: ranked on too few benchmark results · † only the maker's own numbers so far · ~RAM is estimated; full context may need more memory · 830 of 1134 catalogue models listed here · sources and licences
whatmodel is an open-source AI leaderboard database. The data, scoring and ranking engine,
and benchmark tools are available here. This README and pages/ are generated from the
same data and code with npm run pages.
An interactive version of the database can be found at whatmodel.sh. Every model has its own page showing what it's good at, how each test result compares with other models, and how to run it:
| Path | Contents |
|---|---|
data/ | Models, benchmark results, prices, measured speeds and hardware. Every number links to its source. |
data/extracts/ | Share-alike source extracts, published separately as their licences require |
lib/ | The engine: scores (score.ts), memory fit and ranking (fit.ts), page data (views.ts), terminal output (format.ts), markdown pages (markdown.ts) |
pages/ | The other leaderboards, the method and one page per model, generated |
bench.sh, bench.py | Interactive benchmark launcher and runner; see BENCHMARKING.md |
scripts/ | Checks, import, export, page generation and measurement scripts |
licenses/ | Full licence notices for every source |
docs/ | Source audit, validation and coverage reports |
curl "whatmodel.sh/?ram=32&gpu=apple"
curl "whatmodel.sh/image?ram=64&gpu=card&vram=24&format=json"
| Parameter | Meaning |
|---|---|
ram, vram | GB, or raw bytes / kB / MiB as the OS reports them |
gpu | apple, card (NVIDIA, AMD or Intel graphics card) or none |
mac | bytes, optionally followed by a newline and the chip name; implies Apple Silicon |
nv | nvidia-smi --query-gpu=name,memory.total --format=csv,noheader output |
hw | exact chip or card id from data/hardware.json |
max | highest price in the category's unit (LLMs: $ per 1M tokens, blended 3:1 in:out); 0 = local only |
sort, dir | score, where, speed, ram or context; asc or desc |
set | optional model-set filter; omitted = all sets, without changing the overall score scale |
score | set uses only the selected set’s evaluations; omitted = overall scores, with uncalibrated scores/ranks reported as null |
all=1, n=20, format=json, color=1, chart=1 | show models that don't fit, row limit (text: 10, JSON: 40), JSON, ANSI colour, charts |
width=120, help=1 | wider terminal table (default: 80 columns), concise command help |
The same output runs from this repo, with no server:
npm install
npm run board -- "ram=32&gpu=apple"
npm run board -- image "ram=64&gpu=card&vram=24"
Each category's score comes from a one-factor latent model fitted to the benchmark results, within sets
of models that share enough benchmarks to be compared. It lives in lib/score.ts and is explained in
pages/method.md. 50 is an average model, and ± is the uncertainty. npm run check
reports how well the fit predicts results it hasn't seen.
Sponsors never change rankings or scores. Sponsored spots on whatmodel.sh are always labelled, and you can check the ranking against the code here.
data/ directly, or run npm run export to write cleaned snapshots, CSV tables
(scores, models, results, speeds and bench runs) and a SHA-256 manifest to dist/data/. They're the same
files whatmodel.sh serves at whatmodel.sh/data/manifest.json.data/sources.json lists them and
licenses/ has the full notices.The bench measures how fast models run on your machine. It finds the models and AI tools you have and the whatmodel.sh models that fit your memory; you tick what to test, confirm the plan, then review the measurements. It installs or downloads nothing without asking, and offers to set up Python if it's missing. Works on macOS and Linux; on Windows, run it in WSL.
curl -fsSL https://whatmodel.sh/bench.sh | sh
Results stay local until you choose to open a prefilled GitHub issue. Review it and submit it yourself in the browser. API and agent tests may incur provider charges. See BENCHMARKING.md for supported runners, measurement protocols and what the public report contains.
Submitted results are released under CC0 1.0 and reviewed before they're added.
Corrections and new results are welcome, and you don't need to write code. See CONTRIBUTING.md.
Code: MIT. Data: CC BY-SA 4.0. Bench results: CC0 1.0. Copyright 2026 Merlinsbeard Pty Ltd.