Cuda · MIT
mw00/project-maya
Project Maya - run GLM-5.3-Flash (321B MoE), DeepSeek-V4.1-Flash-1017 (748B MoE) and MiniMax-M3.1-Flash (428B MoE) privately on your own GPU(s), from one card to 16, NVIDIA or AMD, Linux or Windows. Engine, server, dashboard, OpenAI- and Anthropic-compatible API, quant tools.
★ 2.5k
Stars
355 forks · 112 open issues
+51/day
Growth · measured
Stars counted between two of our own snapshots roughly a week apart. The repository page gives the exact span.
57.8
Momentum / 100
see the breakdown below
32d ago
Created
last push 2d ago
Star history
How this score was built
| Component | Weight | Earned | Why |
|---|---|---|---|
| Velocity | 55 | 32.9 | +51/day, log-scaled against a 750/day ceiling |
| Acceleration | 15 | 0.0 | last week ran 0.65× the monthly pace |
| Freshness | 15 | 15.0 | last push 2d ago; decays to zero at 180 days |
| Scale | 15 | 9.9 | 2.5k stars, log-scaled against 150k — stops a 3-star repo topping the board |
| Momentum | 100 | 57.8 | sum of the above |
Detail
| Topics | cudadeepseekdeepseek-v4glmglm-5llmllm-inferencelocal-llmminimaxminimax-m3mixture-of-expertsmoeopenai-apiquantizationrocmself-hosted |
| Matched on | LLM & Inference (llm, llm-inference) Dev Tools (self-hosted) |
| Short window | +360 stars over 7 days = 51.4/day |
| Long window | +1.7k stars over 22 days = 78.9/day |
| Lifetime average | 85.64 stars/day over 30 days |
| First seen here | 2026-10-14 |
| Last read | 2d ago |