Open Source LLM Gateway π¦
See the code
AI Agents that don't break
Manifest LLM Gateway is an open-source LLM gateway for AI agents and apps. Connect your API keys, subscriptions, and local models to one OpenAI-compatible endpoint, and each query goes to the right model. No single-provider lock-in.
Go to app.manifest.build and follow the guide.
Manifest ships as a Docker image. One command:
bash <(curl -sSL https://raw.githubusercontent.com/mnfst/llm-gateway/main/docker/install.sh)
Open http://localhost:2099 and sign up β the first account you create becomes the admin. Full self-hosting guide: docker/DOCKER_README.md.
| Platform | Notes |
|---|---|
| Railway | Best path. Template includes Manifest, PostgreSQL, and S3-compatible storage for request recordings. |
| Render | Blueprint includes Manifest, PostgreSQL, and a persistent recording disk. |
| DigitalOcean | App Platform and PostgreSQL; provide a private Space for recordings. |
| AWS | CloudFormation provisions ECS, RDS, and a private recording bucket. |
| GCP | DeployStack provisions Cloud Run, Cloud SQL, and Cloud Storage. |
Every deployment path now uses durable request-recording storage. Railway, AWS, GCP, and Fly.io provision it natively; Render, Coolify, Easypanel, Docker, and Apple Containers mount persistent storage. DigitalOcean, Heroku, and Koyeb collect external S3-compatible settings during setup. Volume-backed templates are single-instance; use S3-compatible storage before scaling horizontally.
Full deployment guides: Railway, Render, DigitalOcean, AWS, GCP, Fly.io, Coolify, Easypanel, Heroku, Koyeb, and Apple Containers.
The old npm-based self-hosting path is no longer supported. Use the Docker image or one of the deployment guides above.
Manifest connects to 300+ models through 35 built-in provider connections plus any custom OpenAI/Anthropic-compatible endpoint. Bring your own API key, reuse one of 18 subscription flows, or run models locally. Everything is routed through the same OpenAI-compatible endpoint β send "model": "auto" and Manifest picks the model.
Provider catalogs are discovered dynamically when credentials are connected. The examples below are representative, not exhaustive.
| Provider | API key / local | Subscription | Model catalog |
|---|---|---|---|
| OpenAI | β | β ChatGPT Plus / Pro / Team | GPT-5.6 (Sol / Terra / Luna), GPT-5.5, GPT-5.4, Codex, o-series |
| Anthropic | β | β Claude Max / Pro | Claude Opus 5, Sonnet 5, Fable 5, Haiku 4.5 |
| β | β Sign in with Google | Gemini 3.5 Flash, 3.1 Flash-Lite, Gemini 2.5 | |
| Google Vertex AI | β | β | Gemini models via Vertex AI |
| Gemini Free | β Managed key | β | Free Gemini models through Manifest's managed gateway |
| Meta | β | β | Muse Spark 1.2 / 1.1 + Contributor route (Meta Model API) |
| xAI | β | β Grok subscription | Grok 4.5, Grok 4.3, Grok Build, Grok 4.20 |
| AWS Bedrock | β | β | Claude, GPT, Kimi, MiniMax, Nemotron, Nova via Bedrock |
| Alibaba Cloud / Qwen | β | β Qwen Token Plan | Qwen 3.7 Max / Plus / Flash, DeepSeek, Kimi, GLM |
| DeepSeek | β | β | DeepSeek V4 Pro, V4 Flash, V3.2, R1 |
| Mistral | β | β Mistral Vibe | Mistral Large, Medium 3.5, Devstral, Codestral |
| Moonshot (Kimi) | β | β Kimi Coding Plan | Kimi K3, K2.7 Code, Kimi for Coding |
| MiniMax | β | β MiniMax Coding Plan | MiniMax M3, M2.7, M2.5 |
| Xiaomi MiMo | β | β MiMo Token Plan | MiMo V2.5 Pro, V2.5, Flash |
| Z.ai | β | β GLM Coding Plan | GLM 5.2, GLM 5.1, GLM 5 Turbo |
| BytePlus | β | β ModelArk Coding Plan | Ark Code, Seed Code, GLM, Kimi, DeepSeek, GPT-OSS |
| GitHub Copilot | β | β Copilot subscription | Claude, GPT, Gemini, Grok via Copilot |
| Kiro | β | β Kiro subscription | kiro/auto, Claude, DeepSeek, MiniMax, GLM, Qwen |
| Command Code | β | β Command Code subscription | Claude, DeepSeek V4, Qwen 3.7, Gemini, Kimi |
| ClinePass | β | β ClinePass subscription | cline-pass/glm-5.2, Kimi, DeepSeek, MiMo, MiniMax, Qwen |
| NousResearch | β | β NousResearch subscription | NousResearch Portal model catalog |
| OpenCode Go | β | β OpenCode Go | DeepSeek V4, Qwen 3.7, GLM, Kimi, MiMo |
| Ollama / Ollama Cloud | π₯οΈ Local | β Ollama Cloud | Local tags: Llama, Qwen, Gemma. Cloud: DeepSeek V4, GLM, Kimi |
| LM Studio | π₯οΈ Local | β | Local GGUF models, port 1234 |
| llama.cpp | π₯οΈ Local | β | Local GGUF models, port 8080 |
| OpenRouter | β | β | 300+ models across labs |
| OpenCode Zen | β | β | Curated Claude, GPT, DeepSeek, MiMo, Nemotron |
| Kilo | β | β | Kilo Gateway catalog |
| Cerebras | β | β | GPT-OSS, GLM, Gemma on Cerebras inference |
| Fireworks AI | β | β | DeepSeek V4, Kimi K2.7, Qwen 3.7, Nemotron |
| Groq | β | β | Llama 4, Qwen 3.6, GPT-OSS, Gemma |
| Hugging Face | β | β | Open models through Hugging Face Inference Providers |
| NVIDIA NIM | β | β | Nemotron 3, GLM, Kimi, MiniMax, Qwen |
| Pioneer | β | β | OpenAI-compatible and fine-tuned Pioneer models |
| Custom | β | β | Any /v1/chat/completions or /v1/messages endpoint |
n8n-nodes-manifest on npm(top 24 of 79)
483 followers Β· starred Mar 2026
655 followers Β· starred Aug 2026
396 followers Β· starred Aug 2026
7,705 followers Β· starred Apr 2025
Open Source LLM Gateway π¦
See the code
AI Agents that don't break
Manifest LLM Gateway is an open-source LLM gateway for AI agents and apps. Connect your API keys, subscriptions, and local models to one OpenAI-compatible endpoint, and each query goes to the right model. No single-provider lock-in.
Go to app.manifest.build and follow the guide.
Manifest ships as a Docker image. One command:
bash <(curl -sSL https://raw.githubusercontent.com/mnfst/llm-gateway/main/docker/install.sh)
Open http://localhost:2099 and sign up β the first account you create becomes the admin. Full self-hosting guide: docker/DOCKER_README.md.
| Platform | Notes |
|---|---|
| Railway | Best path. Template includes Manifest, PostgreSQL, and S3-compatible storage for request recordings. |
| Render | Blueprint includes Manifest, PostgreSQL, and a persistent recording disk. |
| DigitalOcean | App Platform and PostgreSQL; provide a private Space for recordings. |
| AWS | CloudFormation provisions ECS, RDS, and a private recording bucket. |
| GCP | DeployStack provisions Cloud Run, Cloud SQL, and Cloud Storage. |
Every deployment path now uses durable request-recording storage. Railway, AWS, GCP, and Fly.io provision it natively; Render, Coolify, Easypanel, Docker, and Apple Containers mount persistent storage. DigitalOcean, Heroku, and Koyeb collect external S3-compatible settings during setup. Volume-backed templates are single-instance; use S3-compatible storage before scaling horizontally.
Full deployment guides: Railway, Render, DigitalOcean, AWS, GCP, Fly.io, Coolify, Easypanel, Heroku, Koyeb, and Apple Containers.
The old npm-based self-hosting path is no longer supported. Use the Docker image or one of the deployment guides above.
Manifest connects to 300+ models through 35 built-in provider connections plus any custom OpenAI/Anthropic-compatible endpoint. Bring your own API key, reuse one of 18 subscription flows, or run models locally. Everything is routed through the same OpenAI-compatible endpoint β send "model": "auto" and Manifest picks the model.
Provider catalogs are discovered dynamically when credentials are connected. The examples below are representative, not exhaustive.
| Provider | API key / local | Subscription | Model catalog |
|---|---|---|---|
| OpenAI | β | β ChatGPT Plus / Pro / Team | GPT-5.6 (Sol / Terra / Luna), GPT-5.5, GPT-5.4, Codex, o-series |
| Anthropic | β | β Claude Max / Pro | Claude Opus 5, Sonnet 5, Fable 5, Haiku 4.5 |
| β | β Sign in with Google | Gemini 3.5 Flash, 3.1 Flash-Lite, Gemini 2.5 | |
| Google Vertex AI | β | β | Gemini models via Vertex AI |
| Gemini Free | β Managed key | β | Free Gemini models through Manifest's managed gateway |
| Meta | β | β | Muse Spark 1.2 / 1.1 + Contributor route (Meta Model API) |
| xAI | β | β Grok subscription | Grok 4.5, Grok 4.3, Grok Build, Grok 4.20 |
| AWS Bedrock | β | β | Claude, GPT, Kimi, MiniMax, Nemotron, Nova via Bedrock |
| Alibaba Cloud / Qwen | β | β Qwen Token Plan | Qwen 3.7 Max / Plus / Flash, DeepSeek, Kimi, GLM |
| DeepSeek | β | β | DeepSeek V4 Pro, V4 Flash, V3.2, R1 |
| Mistral | β | β Mistral Vibe | Mistral Large, Medium 3.5, Devstral, Codestral |
| Moonshot (Kimi) | β | β Kimi Coding Plan | Kimi K3, K2.7 Code, Kimi for Coding |
| MiniMax | β | β MiniMax Coding Plan | MiniMax M3, M2.7, M2.5 |
| Xiaomi MiMo | β | β MiMo Token Plan | MiMo V2.5 Pro, V2.5, Flash |
| Z.ai | β | β GLM Coding Plan | GLM 5.2, GLM 5.1, GLM 5 Turbo |
| BytePlus | β | β ModelArk Coding Plan | Ark Code, Seed Code, GLM, Kimi, DeepSeek, GPT-OSS |
| GitHub Copilot | β | β Copilot subscription | Claude, GPT, Gemini, Grok via Copilot |
| Kiro | β | β Kiro subscription | kiro/auto, Claude, DeepSeek, MiniMax, GLM, Qwen |
| Command Code | β | β Command Code subscription | Claude, DeepSeek V4, Qwen 3.7, Gemini, Kimi |
| ClinePass | β | β ClinePass subscription | cline-pass/glm-5.2, Kimi, DeepSeek, MiMo, MiniMax, Qwen |
| NousResearch | β | β NousResearch subscription | NousResearch Portal model catalog |
| OpenCode Go | β | β OpenCode Go | DeepSeek V4, Qwen 3.7, GLM, Kimi, MiMo |
| Ollama / Ollama Cloud | π₯οΈ Local | β Ollama Cloud | Local tags: Llama, Qwen, Gemma. Cloud: DeepSeek V4, GLM, Kimi |
| LM Studio | π₯οΈ Local | β | Local GGUF models, port 1234 |
| llama.cpp | π₯οΈ Local | β | Local GGUF models, port 8080 |
| OpenRouter | β | β | 300+ models across labs |
| OpenCode Zen | β | β | Curated Claude, GPT, DeepSeek, MiMo, Nemotron |
| Kilo | β | β | Kilo Gateway catalog |
| Cerebras | β | β | GPT-OSS, GLM, Gemma on Cerebras inference |
| Fireworks AI | β | β | DeepSeek V4, Kimi K2.7, Qwen 3.7, Nemotron |
| Groq | β | β | Llama 4, Qwen 3.6, GPT-OSS, Gemma |
| Hugging Face | β | β | Open models through Hugging Face Inference Providers |
| NVIDIA NIM | β | β | Nemotron 3, GLM, Kimi, MiniMax, Qwen |
| Pioneer | β | β | OpenAI-compatible and fine-tuned Pioneer models |
| Custom | β | β | Any /v1/chat/completions or /v1/messages endpoint |
n8n-nodes-manifest on npm(top 24 of 79)
483 followers Β· starred Mar 2026
655 followers Β· starred Aug 2026
396 followers Β· starred Aug 2026
7,705 followers Β· starred Apr 2025