OSS Radar
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… Why it matters: useful plumbing for shipping LLM-powered features.
GitHub description: 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.
Stars
2k
Last 24h
+34est.
Last 7 days
+235est.
Forks
273
Star history (last 1 snapshot)
Only one snapshot so far. The chart fills in as OSS Radar collects daily snapshots.
est. means the star gain is estimated, not measured: until OSS Radar has collected enough daily snapshots for a repo, we use total stars ÷ days since creation (or extrapolate from the snapshots we do have). Estimates are replaced by measured deltas automatically as history builds up.
Maintainer? Add the badge to your README
Shows your live OSS Radar rank and links back here. Updates hourly.
[](https://ossradar.net/repo/syv-ai/HyperQwen)<a href="https://ossradar.net/repo/syv-ai/HyperQwen"><img src="https://ossradar.net/badge/syv-ai/HyperQwen" alt="Trending on OSS Radar"></a>About HyperQwen
syv-ai/HyperQwen is a Python project in the LLM tooling category, created 59 days ago. It has 2k GitHub stars and 273 forks, and gained +235 stars over the last 7 days (early estimate), ranking #151 of 167 LLM tooling repos tracked by OSS Radar this week.
See more trending LLM tooling repos or the top AI repos this week.
- Language
- Python
- Category
- LLM tooling
- Created
- Aug 15, 2026
- Last push
- Oct 11, 2026
- consumer-gpu
- cuda
- inference-optimization
- kv-cache
- llm-inference
- local-llm
- quantization
- qwen
- qwen3
- rtx-3090
- speculative-decoding
- vllm
Similar rising repos
- StayLameBro/backburner+595est.
Your iPhone helps your Mac run a 27B model: faster prompt reading and more context over a USB-C cable. Why it matters: useful plumbing for shipping LLM-powered features.
- vllm-project/vllm+489est.
A high-throughput and memory-efficient inference and serving engine for LLMs. Why it matters: useful plumbing for shipping LLM-powered features.
- datawhalechina/zero-to-sglang+216est.
Official SGLang x Datawhale course on LLM inference: understand inference, build a mini-sglang from scratch, then read the real SGLang source and… Why it matters: useful plumbing for shipping LLM-powered features.
- bojieli/ai-infra-book+863est.
《深入理解 AI Infra：量化分析与系统设计》（李博杰 著）开源书稿：从硬件约束和模型架构出发，量化推导 LLM 推理与训练系统设计。含全书正文、PDF、配套计算工具与实验. Why it matters: useful plumbing for shipping LLM-powered features.
- hiyouga/LlamaFactory+428est.
Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024). Why it matters: useful plumbing for shipping LLM-powered features.
- MakazhanAlpamys/Soup+259est.
Fine-tune LLMs from one YAML. Layer streaming trains an 8B model on a 4 GB laptop GPU. Why it matters: useful plumbing for shipping LLM-powered features.