Cuda · Apache-2.0
yalun753/moe-l2
MoE expert offload for low-VRAM GPUs — run 100B+ MoE models (DeepSeek, Qwen, Mixtral) on 8 GB cards. Expert cache with LRU hot experts, OpenAI-compatible proxy, GGUF multi-shard.
★ 145
Stars
6 forks · 2 open issues
−1.0/wk
Growth · measured
Stars counted between two of our own snapshots roughly a week apart. The repository page gives the exact span.
18.4
Momentum / 100
see the breakdown below
3mo ago
Created
last push 37d ago
Star history
How this score was built
| Component | Weight | Earned | Why |
|---|---|---|---|
| Velocity | 55 | 0.0 | −1.0/wk, log-scaled against a 750/day ceiling |
| Acceleration | 15 | 0.0 | needs 7-day and 30-day rates side by side |
| Freshness | 15 | 12.1 | last push 37d ago; decays to zero at 180 days |
| Scale | 15 | 6.3 | 145 stars, log-scaled against 150k — stops a 3-star repo topping the board |
| Momentum | 100 | 18.4 | sum of the above |
Detail
| Topics | consumer-gpudeepseekexpert-cachingllama-cppllm-inferencemoevram-optimization |
| Matched on | LLM & Inference (llm-inference) |
| Short window | +-1 stars over 7 days = -0.1/day |
| Long window | +-1 stars over 26 days = 0.0/day |
| Lifetime average | 1.42 stars/day over 102 days |
| First seen here | 2026-10-11 |
| Last read | 2d ago |