Python
peonist-ai/halogen-server
The fastest way to run Qwen3.8 27B on Strix Halo (gfx1151)
★ 157
Stars
8 forks · 6 open issues
+2.0/day
Growth · measured
Stars counted between two of our own snapshots roughly a week apart. The repository page gives the exact span.
35.6
Momentum / 100
see the breakdown below
2mo ago
Created
last push 54d ago
Star history
How this score was built
| Component | Weight | Earned | Why |
|---|---|---|---|
| Velocity | 55 | 9.1 | +2.0/day, log-scaled against a 750/day ceiling |
| Acceleration | 15 | 9.4 | last week ran 1.63× the monthly pace |
| Freshness | 15 | 10.7 | last push 54d ago; decays to zero at 180 days |
| Scale | 15 | 6.4 | 157 stars, log-scaled against 150k — stops a 3-star repo topping the board |
| Momentum | 100 | 35.6 | sum of the above |
Detail
| Topics | amdgfx1151gpu-inferencehipinference-enginellmllm-servinglocal-llmopenai-apiqwenqwen3radeonrocmryzen-aispeculative-decodingstrix-halo |
| Matched on | LLM & Inference (llm-serving) |
| Short window | +14 stars over 7 days = 2.0/day |
| Long window | +32 stars over 26 days = 1.2/day |
| Lifetime average | 2.20 stars/day over 71 days |
| First seen here | 2026-10-11 |
| Last read | 2d ago |