B-I-T-W-I-S-E-M-I-N-D-S/-moshi-ditto-streaming-debug-v1-moshi-8-w-v2

Initially, streaming works well (Moshiq8 model + H200 GPU).

0

stars

4

commits

Python

primary language

Apr 12, 2026

updated

README

-moshi-ditto-streaming-debug-v1-moshi-8-w-v2

Contributors

3 commits

1 commits

Similar repos

utsaa/personaplex-work-multi-gpu

Personaplex work with multi gpu

0

tensorforger/FluxRT

Real-time stream editing pipeline powered by the FLUX.2-klein-4B model, optimized for consumer GPUs

445

christopher-kapic/MiMo-V2.5-ROCmFP4-GGUF

3

xltzsoft/deepseek-v4-sm89

面向 8×RTX 4090（SM89）的 DeepSeek-V4-Flash-0731 MXFP4 SGLang 实验分支

7

animede/Diffusers_minimax-h3

MiniMax-H3 (video + synchronized audio, 33B) on diffusers — 9.7s per still, 44.2s per 5s clip on…

19

Luce-Org/lucebox

LLM speculative inference server for heterogeneous hardware & consumer GPUs

2,843

Luce-Org/lucebox-hub

LLM speculative inference server for heterogeneous hardware & consumer GPUs

2,843

julianmb/q38rocm

Qwen 3.8 27B ROCmFP4 on AMD Strix Halo (Ryzen AI Max+ 395). Up to 36 tok/s via MTP Speculation,…

205

Languages

Python

88.0%

Jupyter Notebook

6.6%

HTML

4.4%

B-I-T-W-I-S-E-M-I-N-D-S/-moshi-ditto-streaming-debug-v1-moshi-8-w-v2

Initially, streaming works well (Moshiq8 model + H200 GPU).

0

stars

4

commits

Python

primary language

Apr 12, 2026

updated

README

-moshi-ditto-streaming-debug-v1-moshi-8-w-v2

Similar repos

utsaa/personaplex-work-multi-gpu

Personaplex work with multi gpu

0

tensorforger/FluxRT

Real-time stream editing pipeline powered by the FLUX.2-klein-4B model, optimized for consumer GPUs

445

christopher-kapic/MiMo-V2.5-ROCmFP4-GGUF

3

xltzsoft/deepseek-v4-sm89

面向 8×RTX 4090（SM89）的 DeepSeek-V4-Flash-0731 MXFP4 SGLang 实验分支

7

animede/Diffusers_minimax-h3

MiniMax-H3 (video + synchronized audio, 33B) on diffusers — 9.7s per still, 44.2s per 5s clip on…

19

Luce-Org/lucebox

LLM speculative inference server for heterogeneous hardware & consumer GPUs

2,843

Luce-Org/lucebox-hub

LLM speculative inference server for heterogeneous hardware & consumer GPUs

2,843

julianmb/q38rocm

Qwen 3.8 27B ROCmFP4 on AMD Strix Halo (Ryzen AI Max+ 395). Up to 36 tok/s via MTP Speculation,…

205

Contributors

3 commits

1 commits

Languages

Python

88.0%

Jupyter Notebook

6.6%

HTML

4.4%