Shell
peonist-ai/halogen-flash-server
Local inference that behaves like a frontier API: Qwen3.8-Flash-Next on AMD Strix Halo plus small NPU models, over the OpenAI and Anthropic APIs on one port.
★ 1.2k
Stars
81 forks · 30 open issues
+13/day
Growth · measured
Stars counted between two of our own snapshots roughly a week apart. The repository page gives the exact span.
48.7
Momentum / 100
see the breakdown below
2mo ago
Created
last push 4d ago
Star history
How this score was built
| Component | Weight | Earned | Why |
|---|---|---|---|
| Velocity | 55 | 22.2 | +13/day, log-scaled against a 750/day ceiling |
| Acceleration | 15 | 2.7 | last week ran 1.18× the monthly pace |
| Freshness | 15 | 14.9 | last push 4d ago; decays to zero at 180 days |
| Scale | 15 | 9.0 | 1.2k stars, log-scaled against 150k — stops a 3-star repo topping the board |
| Momentum | 100 | 48.7 | sum of the above |
Detail
| Topics | amdgfx1151gpu-inferencehipinference-enginellmllm-servinglocal-llmlong-contextmixture-of-expertsopenai-apiqwenqwen3radeonrocmryzen-aispeculative-decodingstrix-halo |
| Matched on | LLM & Inference (llm-serving) |
| Short window | +94 stars over 7 days = 13.4/day |
| Long window | +296 stars over 26 days = 11.4/day |
| Lifetime average | 17.32 stars/day over 71 days |
| First seen here | 2026-10-11 |
| Last read | 2d ago |