Hackernews posts about Qwen 3

1. RTX 5080 and RTX 3090 Setup: 80 Tok/s on Qwen 3.6 27B Q8 (imil.net)
   291 points by iMil 8 days ago | 108 comments
2. Rio de Janeiro's city government model Rio3.5 beats Qwen3.7 in recent benchmarks (twitter.com)
   142 points by lucasfcosta 7 days ago | 45 comments
3. Two Qwen3 models on one DGX Spark: the residency math (www.devashish.me)
   74 points by devashish86 3 days ago | 34 comments
4. Qwen3.7-Plus: Multimodal Agent Intelligence (qwen.ai)
   42 points by meetpateltech 20 days ago | 12 comments
5. Qwen3.7-Max Ran for 35 Hours on Unknown Hardware and Achieved a 10× Speedup (firethering.com)
   38 points by steveharing1 27 days ago | 31 comments
6. Running Claude Code Offline on an M3 Pro with Qwen3.6 (har-ki.github.io)
   18 points by har-ki 10 days ago | 10 comments
7. Show HN: FlashQwen – A from-scratch CUDA inference engine for Qwen3 (github.com)
   5 points by langtang1996 6 days ago | discuss
8. Qwen 3.7 Plus (artificialanalysis.ai)
   4 points by theanonymousone 18 days ago | 1 comments
9. Qwen 3.6 93B with MTP on 2×RTX 3090 NVLink=187 tokens/SEC,LLM lost bleat-a-thon (github.com)
   4 points by devilfileprong 7 days ago | discuss
10. Tuning CPU-only Qwen3-30B inference with an IBM Quantum sampling loop (github.com)
    2 points by Royce-CMR 23 days ago | discuss
11. Launch HN: General Instinct (YC P26) – Frontier models on edge devices
    63 points by guanming0717 16 days ago | 16 comments
12. Show HN: Learn from 30 historical figures, open source, nonprofit, self-hosted (github.com)
    49 points by micstradev 12 days ago | 24 comments
13. Show HN: Hitoku Draft – Context aware local assistant (hitoku.me)
    21 points by lostathome 17 days ago | 8 comments
14. I Replaced My AI Agent's Flat Fact Store with a Graph Database
    10 points by grawl_dorgiers 18 days ago | 4 comments
15. Show HN: I am running 3 coding agents non-stop over the last 3 days. Here is how
    10 points by sermakarevich 8 days ago | 3 comments
16. Ask HN: Any Local LLM can I run without GPU for Local Agentic workflow AI?
    8 points by limondas 11 days ago | 4 comments
17. Show HN: Selora – local model for Home Assistant (github.com)
    7 points by bayshark 4 days ago | 4 comments
18. Story of How Im Running an Unlimited $6/Month AI Provider on 4x RTX 3090s
    7 points by yolo-auto 8 days ago | 3 comments
19. Show HN: Levi – run AlphaEvolve on your Claude Code/Codex for dirt cheap (ttanv.github.io)
    5 points by ttanv 13 days ago | discuss
20. Show HN: Free open source coding models in Slack (www.runcord.com)
    3 points by ramonga 24 days ago | discuss
21. Ask HN: What are some good/fast coding models for Apple Silicon?
    2 points by LoganDark 5 days ago | 3 comments
22. Show HN: Best setup local LLM found for a 5090 (llama.cpp fork + turboquant) (local-llm.utop.workers.dev)
    2 points by utopman 14 days ago | 2 comments
23. Why aren't languages/frameworks offering retrained models for their project?
    2 points by ckluis 14 days ago | 1 comments
24. Show HN: Clipto-fully local, natural language search over terabytes of media (www.clipto.com)
    2 points by henry_kang 19 days ago | 1 comments
25. Ask HN: What do you do when you hit Claude subscription limits?
    2 points by sermakarevich 9 days ago | discuss
26. The OnlyFans Economy of American AI (leoveanu.com)
    146 points by futurisold 14 days ago | 204 comments
27. A Comprehensive Analysis of the Open-Weight Frontier (May 2026) (deepresearch.ninja)
    2 points by jackalxyz 25 days ago | discuss
28. DocumentAI Visual Benchmark - GPT 5.5, Gemini 3.5, Qwen... (www.maltebuettner.eu)
    2 points by hegelsmind 22 days ago | discuss
29. Qwen and Fable: An open-weights agentic coding model. 35B Mixture-of-Experts (huggingface.co)
    2 points by thunderbong 4 days ago | discuss
30. Ask HN: Is it feasible to run a model on device for complete privacy?
    3 points by mazinz 14 days ago | 6 comments