Python · Apache-2.0
syv-ai/HyperQwen
Serve large Qwen models fast on the GPUs you actually own. Qwen3.8-27B on a single 24 GB card with vLLM: 127 tok/s single-user (381 when the answer quotes the prompt), ~1,035 tok/s at 64 concurrent, 150k-262k context. vLLM patches, requant pipeline, benchmarks.
★ 2.6k
Stars
324 forks · 6 open issues
+23/day
Growth · measured
Stars counted between two of our own snapshots roughly a week apart. The repository page gives the exact span.
51.2
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 | 26.5 | +23/day, log-scaled against a 750/day ceiling |
| Acceleration | 15 | 0.0 | last week ran 0.98× the monthly pace |
| Freshness | 15 | 14.9 | last push 4d ago; decays to zero at 180 days |
| Scale | 15 | 9.9 | 2.6k stars, log-scaled against 150k — stops a 3-star repo topping the board |
| Momentum | 100 | 51.2 | sum of the above |
Detail
| Topics | consumer-gpucudainference-optimizationkv-cachellm-inferencelocal-llmquantizationqwenqwen3rtx-3090speculative-decodingvllm |
| Matched on | LLM & Inference (llm-inference) |
| Short window | +163 stars over 7 days = 23.3/day |
| Long window | +618 stars over 26 days = 23.8/day |
| Lifetime average | 31.21 stars/day over 82 days |
| First seen here | 2026-10-11 |
| Last read | 2d ago |