Swift
NeelM0906/Mference
Swift + Metal MoE inference for Apple Silicon: Qwen 3.6 35B at 23.5–29.3 tok/s decode with 2.20× faster long-prompt prefill on a 24 GB M5; Gemma 4 26B in ~2 GB, DeepSeek-V4-Flash 284B, Inkling-Small 276B, native Mac app, CLI, and OpenAI-compatible server.
★ 122
Stars
18 forks · 4 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.
22.1
Momentum / 100
see the breakdown below
3mo ago
Created
last push 4d ago
Star history
How this score was built
| Component | Weight | Earned | Why |
|---|---|---|---|
| Velocity | 55 | 1.1 | +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 | 14.9 | last push 4d ago; decays to zero at 180 days |
| Scale | 15 | 6.1 | 122 stars, log-scaled against 150k — stops a 3-star repo topping the board |
| Momentum | 100 | 22.1 | sum of the above |
Detail
| Topics | apple-silicondeepseekgemmallm-inferencemacosmetalmixture-of-expertsqwenswift |
| Matched on | LLM & Inference (llm-inference) |
| Short window | +1 stars over 7 days = 0.1/day |
| Long window | +0 stars over 26 days = 0.0/day |
| Lifetime average | 1.26 stars/day over 97 days |
| First seen here | 2026-10-11 |
| Last read | 2d ago |