Qwen

What hackers pick instead

• Qwen 3.8 27BAn open-weights model that runs on a single high-end GPU and nears frontier coding quality, though it overthinks, burns tokens and struggles with complex tasks.1173recommend
• Z.ai GLMAn open-weight coding model close to frontier quality at a fraction of the cost, though critics cite slow, outage-prone hosting, erratic output and expensive self-hosting hardware.1043recommend
• KimiMoonshot's open-weight models rival top closed models at coding and research for a fraction of the price, though they can be slow and overthink.931recommend
• Gemma 4Google's open-weight model family runs well on local hardware with strong quality and token efficiency for its size, though critics cite unreliable tool calling and weaker coding than Qwen.713recommend
• Qwen3.6-35B-A3BA fast mixture-of-experts open model that runs well on consumer hardware and is surprisingly capable for its size, though it falls well short of frontier models on complex tasks.373recommend

Every quote

The 40 most detailed of 775 comments. Each links to the original.

1. The Qwen models are cool, but if you're coming from Opus you will be somewhere between mildly to very disappointed depending on the complexity of your work.

   Mixed · Apr 2026 · in “Changes to GitHub Copilot Individual Plans”

2. several Chinese models (at least Qwen, DeepSeek, Kimi and GLM) are "close enough" that with the right harness they are cost effective alternatives

   Recommends · Jun 2026 · in “Uber's $1,500/Month AI Limit Is a Useful Signal for AI Tool Pricing”

3. they do a lot of benchmaxxing after using them. Even at full precision they're never as good as models with similar benchmarks.

   Mixed · Aug 2026 · in “Qwen3.8-27B”

4. By properly structuring my project, I can point it at any extension I want and get it going for 30 minutes to extend whatever.

   Recommends · Jun 2026 · in “What it feels like to work with Mythos”

5. Look at qwen, gemma, etc, they can all do structured output, and they can do it locally and cheaper and/or faster than Jev!

   Recommends · Oct 2026 · in “Jev meets the Trolley Problem”

6. My 64GB DDR4 laptop can already run Qwen 47B at .7 tok/s, that's already usable for some usecases (overnight stuff mostly).

   Mixed · May 2026 · in “Google Declaring War on the Web”

7. DeepSeek, Alibaba/Qwen, Baidu/ERNIE, and Tencent/Hunyuan are breaking that model by releasing various open weight models

   Recommends · May 2026 · in “Greg Brockman: Inside the 72 Hours That Almost Killed OpenAI”

8. It is much better than I thought it would be. But also still falls short in some obvious ways compared to the frontiers.

   Mixed · Mar 2026 · in “What if AI doesn't need more RAM but better math?”

9. publish it to the public, some of not majority contribute with open-weight models making a future of local llm possible

   Recommends · Jan 2026 · in “Kimi Released Kimi K2.5, Open-Source Visual SOTA-Agentic Model”

10. I am feeling pretty liberated with my local Qwen, it can make copies of software, reverse engineer hardware bootloaders

    Recommends · Oct 2026 · in “No Man Is an Island”

11. Alibaba's Qwen is definitely competitive, and its the main story these days if you want to run a coding model locally.

    Recommends · Jan 2026 · in “The recurring dream of replacing developers”

12. They're beating the benchmarks like top-frontier models, focused on open-source and much cheaper than the competitors.

    Recommends · Sep 2026 · in “Qwen Image 2.1”

13. Qwen models are the most recommended ones that will actually run on local hardware that is accessible to the masses

    Recommends · Jun 2026 · in “Ornith-1.0: self-improving open-source models for agentic coding”

14. For anything longer form though, particularly with larger code contexts, Qwen is far more useful for me personally.

    Recommends · May 2026 · in “LFM2-24B-A2B: Scaling Up the LFM2 Architecture”

15. Things that were Qwen or Gemma Diffusion did better at those kinds of tasks but were generally pretty inconsistent

    Mixed · Sep 2026 · in “Kev: Tiny Jev-like family of decision models built on top of Qwen3.5”

16. What's wrong with a country hosting a Kimi, Qwen or GPT-Oss on their hardware for their government work purpose?

    Recommends · Jun 2026 · in “GPT‑NL: a sovereign language model for the Netherlands”

17. I can run a quantised version of Gemma or Qwen on my 16GB M1 Macbook Pro that beats GPT-4 from 2023 hands-down

    Recommends · Jun 2026 · in “Semgrep: GLM 5.2 beats Claude in our Cyber Benchmarks”

18. The only decent model I can find which can compete sort of with Cloud models without costing a bank in GPU/RAM

    Recommends · Feb 2026 · in “Time Is Different”

19. it runs Qwen surprisingly well and I didn't and don't care too much about that except to poke it occasionally

    Mixed · Sep 2026 · in “Nvidia is the central bank of AI”

20. I'd stick to the smaller 7b-18b parameter range or MOE models like Qwen if you want a usable inference speed.

    Recommends · Apr 2026 · in “How good is Mac Studio M3 Ultra for Trillion param models like DeepSeekv4?”

21. Qwen is also censored - although since it's open weight, there are completely uncensored versions available.

    Mixed · Jun 2026 · in “Claude Sonnet 5”

22. it feels a bit like last year's Sonnet. They don't quite make the leaps of understanding you get from Opus.

    Mixed · Apr 2026 · in “Anthropic no longer allowing Claude Code subscriptions to be used with OpenClaw”

23. Local models can quickly read and explain codebases, even if they can't write them - this is a superpower

    Recommends · Jun 2026 · in “Local Qwen isn't a worse Opus, it's a different tool”

24. Multiple ground up initiatives here in Rwanda would not be possible without models such as Qwen or NLLB.

    Recommends · May 2026 · in “Open-Weight Models Don't Need to Win”

25. qwen is very good at ALMOST getting the job done for large tasks and does fine on smaller/medium tasks.

    Mixed · Jul 2026 · in “Jamesob's guide to running SOTA LLMs locally”

26. We already have censorship and state values in Chinese models (and have for awhile, ask Qwen about “sensitive” issues like Taiwan)

    Criticizes · May 2026 · in “Teaching Claude Why”

27. the Qwen model series considers it highly unethical to ever talk about Taiwan as anything other than a renegade province of China

    Criticizes · May 2026 · in “Teaching Claude Why”

28. it started with “your premise is incorrect bla bla bla” and then it just redacted the whole thing and showed me an error code.

    Criticizes · Jul 2026 · in “Qwen3.8 is launching and going open-weight soon”

29. Qwen and GLM both promise the stars in the sky every single release and the results are always firmly in the "whatever" range

    Criticizes · Feb 2026 · in “GLM-5: From Vibe Coding to Agentic Engineering”

30. on my setup I found it a bit slow for my use case, and the benefits weren't strong enough to outweigh the slowdown

    Criticizes · Sep 2026 · in “Lichen – A local, BOY model system1 (Jev) server with image support”

31. it finds the answer relatively quickly but then second guesses itself multiple times for another 20,000+ tokens

    Criticizes · Aug 2026 · in “Meta Muse Glimmer – open weights 30B local coding model”

32. Maybe it can't work in a hands-off/one-shot fashion like the frontier models, but they can help tremendously if you already have some domain knowledge.

    Recommends · Jul 2026 · in “Who manages the agents?”

33. you could also consider it a miracle that the qwen models are able to perform so well when being trained on inefficient wrapper code data

    Mixed · May 2026 · in “Accelerating Gemma 4: faster inference with multi-token prediction drafters”

34. If you have the RAM and you use models like Qwen4 (3.8 flash next) that can fit 6 to 14 262k contexts in 48GB of system ram as cache.

    Recommends · Sep 2026 · in “AI companies in race to demonstrate their model most threatening to humanity”

35. a couple of other results, both Qwen models, are skewed upward in the detect % ranking because of failure to complete all cases

    Mixed · Jun 2026 · in “Will It Mythos?”

36. Qwen and the OpenAI models are great, aggressive agents, but they're not as good as the anthropic models for human interaction.

    Mixed · Aug 2026 · in “Qwen3.8 Max now ranked as the best overall model by agentic index”

37. I have some chats on research and code architecture that have stretched for weeks without any noteable change in quality

    Recommends · Sep 2026 · in “MiMo v2.6”

38. Qwen's models have clearly been distilled from Anthropic models because they have a much closer communication style

    Recommends · May 2026 · in “Cursor Introduces Composer 2.5”

39. I use QWEN and other local models all the time, but for more clearly defined monitoring and classification tasks.

    Mixed · May 2026 · in “Was my $48K GPU server worth it?”

40. it was good enough that it was able to complete (sometimes) a full job application flow for given json user data

    Recommends · Jul 2026 · in “Weave Robotics launches Isaac 1, a $7,999 home robot with fall 2026 deliveries”