hyperqwen
hyperqwen is 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.... It is ranked #146 on the Dev Radar, in Hosting & infra, first seen 2 days ago and shared in 2 posts (51.2k views).
GitHub - 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. · GitHub Skip to content Navigation Menu Sign in Search / Sign in Sign up Appearance settings You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window. Reload to refresh your session. Dismiss alert {{ message }} Repository…
What people said about hyperqwen on X
1 RTX 3090 now serves Qwen3.8-27B to 64 chats at once. 1035 tok/s total. 127 tok/s for a single chat. Up to 150K context. Card capped at 250W. Setup is 2 commands in docker compose. I run 4 of these cards. On just 1 of them, Qwen3.8 Flash-Next does 87 tok/s for me. The card is 6 years old. The model is open. People…
— @superalesha, 1 days ago · 653 likes · see the post
Tenho mostrado Qwen+Basalt na 5090, mas pra quem tem 4090 ou 3090 parece que esta opção pode ser boa: https://github.com/syv-ai/HyperQwen Obviamente não ter performance de 5090, mas pode ajudar.
— @AkitaOnRails, 2 days ago · 60 likes · see the post
Alternatives to hyperqwen
- boat by ASCII — omfg @theo just released a video that talks about
- Cloudflare Docs — Deploy it to Cloudflare
- Cloudflare Blog — Cloudflare has patched the issue
- Railway — Connect Neon
- Giotto OS — Use Giotto
- Boundless — Apply for access
hyperqwen in numbers
- Rank on the Dev Radar: #146 of 8798
- Shared in 2 posts by 2 accounts: @superalesha, @AkitaOnRails
- 51.2k views on those posts
- First seen 2 days ago, last shared 1 days ago
- Pricing seen by AGK Intelligence: open source
- Market: Hosting & infra
FAQ
What is 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... It was first shared on X 2 days ago and is ranked #146 on the Dev Radar.
Is hyperqwen free?
It is open source.
Who shared hyperqwen?
2 accounts on X, including @superalesha, @AkitaOnRails, in 2 posts totalling 51.2k views.
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 31.2k posts from 5k X accounts over the last 21 days, 8.8k tools, 12 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-10-30 20:27 UTC. Full method.