Qwen 3.8 27B: Open Weights, Best Local Dense Model Yet
Qwen 3.8 27B is out: open weights, best local dense model yet
Qwen releases Qwen3.8-27B, a 27B-parameter dense model with open weights, claiming it's the most capable open model yet. It excels in coding, agentic tasks, and multimodal understanding, with native vision-language support and flexible thinking control. The FP8 version offers near-identical performance and is compatible with major frameworks. Benchmarks show it outperforming larger models like Qwen3.7-Plus and Muse Glimmer-30B in many tasks.
Qwen3.8-27B brings these advances to a compact, deployment-friendly dense model: a native vision-language model that understands images and videos, with flexible thinking control, designed to carry complex, multi-step tasks through to completion with greater reliability.
- CMay
Credit where it's due. Qwen 3.8 27B is only the second local model after Gemma 4 that managed to correctly reason through one of my private benchmarks. It took 5x as many tokens to do it and 12m30s with MTP enabled, but it did do it.
Gemma 4 reasoned through it more implicitly, while Qwen 3.8 reasoned more explicitly. Laguna and Muse Glimmer failed hard on it, though they're useful for other tasks.
The VRAM usage seems way less efficient than Gemma 4 or Glimmer though, with 32K of context taking 2.5GB of VRAM. With those, even with MTP or a DFlash model loaded, you could still fit 256k-768k of context. With Qwen 3.8 27B I can't even fit 128k if I quantize V to Q4_0. Maybe with some trial and error I can find some settings that perform well enough with a larger context window that it's still useful for longer tasks.
Lots more testing to do, though I was getting some decent results out of Muse Glimmer which was more than twice as fast and supported huge context windows, managing to solve some bugs that Gemma 4 struggled with. I can't even begin to throw that task at Qwen, because just the prompt alone would use the entire context window and then it would reason for probably that same amount.
If you've got a 32GB card, it should be a decent model even if it really is memory hungry.
- simonw
Absolutely the best pelican I've seen from a model that runs on my laptop: https://tools.simonwillison.net/markdown-svg-renderer#url=ht...
Bicycle is the right shape. Pelican beak is excellent. Nice background. Most importantly, the pelican has one leg on each side of the bicycle - that's very rare.
(No chain on this bicycle though - in the reasoning trace it says "already chainstay... skip chain detail; maybe a small chainring.")
I ran that on an M5 Max MacBook Pro using LM Studio and their 17GB GGUF: https://lmstudio.ai/models/qwen3.8
It took 21 minutes(!) and used 22,276 reasoning tokens to produce 3,223 tokens of output.
(For the "they're training on your benchmark now" crowd, all of that cheating didn't prevent it from spending 20 minutes thinking about the task first! You can see the reasoning trace in the link I shared.)
For comparison, here's one I got from qwen3.8-2.4t-a95b on OpenRouter, which is pleasingly animated: https://tools.simonwillison.net/markdown-svg-renderer#url=ht...
- dofm
There's a real change (compared to 3.6) in the way it writes in thinking — it drops words like "to" and "we" in "We need to", talks generally in note form, drops the/and all over the place, avoids "for".
"Need be helpful concise", "Need maybe not overdo", "Need ask!" Almost caveman.
I have an (unsourced, vague) suspicion that this rather unique thinking trace pattern is actually hobbling the MTP predictions, which seem to perform poorly.
Other notes: it uses the trick of repeating the prompt in the thinking trace.
It also worries about hidden chain of thought appearing in the final answer. It talks about "desired oververbosity 9", which is new. A bit GPT-ish.
It is being extraordinarily thorough in thinking through one of my code requests, but I don't know if the net result will be any better than the 35B MoE.
I asked it to ask me clarifying questions — it did, and it offered me a list of defaults I could simply agree to.
I don't think it is necessarily overthinking in the looping sense, but it is in the being exhaustive sense. I need to explore how it does with a tighter reasoning budget.
I am impressed but I am definitely in Camp Please-35B-A3B-When? here, because on an M1 Max this isn't really practical. I hope they do one, though I think they may not.
- satvikpendem
As usual, the Jinja templates are messed up so use this [0] to reduce or turn off thinking, fix tool calling, keep a 100% KV cache hit rate, etc.
[0] https://huggingface.co/froggeric/Qwen-Fixed-Chat-Templates
- onlyrealcuzzo
If the benchmarks don't lie, this is getting very close to Opus 4.6 capability - which was the turning point for me for when AI was "good enough" that it became very hard to justify not using it.
I'm sure there's some benchmaxxing going on, and some things you get only with a a larger model.
But I'm feeling pretty confident if not by Gemma 5 than by mid 2028 we'll have local models that are almost always as good as Opus 4.6 was and in many cases far better.
- Casteil
One thing a lot of people don't seem to factor when hyping Qwen is how much models like this tend to 'overthink' with seemingly endless 'second guessing'. 3.8 seems no different from what I've tried thus far.
As capable as it is, it's hard to justify using it when a competing model (e.g. Gemma4:26b-a3b) can consistently achieve the same or similar response with only 1/10th as many 'thinking' tokens, achieve much higher tokens/second, and take a small fraction of the time. I suppose 'YMMV' depending on your use case.
Also, I haven't used it enough yet to see if it's prone to infinite looping, but its predecessors sure were.
- jjcm
Image->html test for this.
Original images: https://image.non.io/neonRamenDesigns.webp
Qwen 3.8 build: https://html.non.io/neonRamenQwen3.8-27b
Overall I'm very impressed with how well this did. It's a big improvement over 3.6, and it feels on-par with some much, much larger models. I think this one is on-par with Gemini 3.7 Flash.
One thing to note - the build for this on my RTX 6000 pro blackwell took a long time. Easily one of the longest builds I've done. It took around 2 hours to build the site. Obviously we'll have some quants for this soon that will accelerate things, but I was still surprised with how long it took.
Comparison builds from this week:
https://html.non.io/neonRamenGemini3.7
https://html.non.io/neonRamenGLM5.3 (note: non-multimodal)
- LeBit
- ramon156
People will claim it's not comparable to Opus despite it beating the score. I'm not sure I disagree, but I'm also unsure whether I care. Most new models nowadays are "good enough". I cannot complain because I'd rather spend that time improving my prompts and docs. Opus might be a _slight bit better_ at picking up vague hints, but it's also extremely expensive, and I hit the 5 hour limit way too quick.
I care a lot about speed and efficiency right now. For my setup I would like to have 2-3 different model families. I've settled on GLM-5.3 (formerly Deepseek v4 pro 0813) for architecting, Deepseek V4 Pro 0813 for developing, and Gemini flash lite (any recent cheap model) for repo scouting. I'll add another one in the mix for reviewing (in this case Gemini 3.7) and that's all I need.
I've tried most models except Grok.
Qwen is too expensive IMO (Alibaba Cloud subscriptions are hard to come by and I'm not spending 50 euros a month for a tool, so 18 euros it is). If it ever becomes efficient enough to run locally I will definitely look back.
Claude is slow and expensive (the cache hit prices are absurd).
OAI is pretty good, I might add it to my arsenal seeing how cheap it is.
These opinions change every day. Last week I would've never picked Deepseek until I read about the pricing. even post aug 16 it's worth it (although it's getting close to gemini pricing).
Right now my costs are 12 euros a month (z.ai) + whatever deepseek consumes. This typically isn't more than 8 euros a wee […]
- KronisLV
I hope really badly that we'll get a new 35B A3B or similar MoE model!
I also miss the Qwen 3 Coder Next, which was 80B A3B, there are quite a few use cases where a non-dense model <100B would be the sweet spot (when you have the VRAM but not the TDP or compute power). Heck, I'd gladly take A5B or A8B or even A10B as a sort of middle ground.
Also alternate link for viewing the images without signing in: https://xcancel.com/Alibaba_Qwen/status/2088280182356611304