Qwen 3.7 Preview

I am very interested in seeing new qwen models. Qwen3.6 27b is the first one that can do things and doesnt constantly loose "it's mind" and that can be run on a 3090 with a good context size. But it's sometimes getting into a loop.

Re: Qwen 3.7 Preview

#5

Post by giancarlostoro » Mon, May 18, 2026, 5:26 PM UTC

There I was waiting on a smaller version of Qwen 3.6 to drop so I can run it on my Mac, and then bam, they drop this.

Re: Qwen 3.7 Preview

#6

Post by giancarlostoro » Mon, May 18, 2026, 5:27 PM UTC

rspoerri wrote: post #4

I am very interested in seeing new qwen models. Qwen3.6 27b is the first one that can do things and doesnt constantly loose "it's mind" and that can be run on a 3090 with a good context size. But it's sometimes getting into a loop.

I had a flavor of an older version of Qwen (I forget which one to be fair) that was coding along, then lost itself in a loop, I was so confused, it was just a random greenfield "lets see how it does" type of project anyway.

Re: Qwen 3.7 Preview

#7

Post by hydra-f » Mon, May 18, 2026, 5:30 PM UTC

Vision has become totally underappreciated, whereas I believe it brings important advantages to a model

Also, a big caveat in using Qwen models has always been its speech patterns. I do wonder how Google made the Gemma lineup so good at this

Let's hope Alibaba continues to open source its models

Re: Qwen 3.7 Preview

#8

Post by sleepyeldrazi » Mon, May 18, 2026, 5:32 PM UTC

I don't think I can handle another small model release by qwen, I'm still trying to find the limits of 3.6 27B and they are already threatening us with a new one?

But jokes aside, I love the fast iteration, these are most probably again finetunes on the 3.5 architecture that appear better in internal testing, which is still very nice to see. Putting more and more pressure on the bigger labs to perform better is always a good thing.

Re: Qwen 3.7 Preview

#9

Post by jwr » Mon, May 18, 2026, 5:35 PM UTC

hydra-f wrote: post #7

Vision has become totally underappreciated, whereas I believe it brings important advantages to a model Also, a big caveat in using Qwen models has always been its speech patterns. I do wonder how Google made the Gemma lineup so good at this Let's hope Alibaba continues to open source its models

Agreed. Incidentally, in my testing, qwen models (qwen3.6-35b-a3b and earlier 3.5) are WAY better with vision than gemma4-26b-a4b. I would normally want to stick with gemma4 only (I use it for spam filtering), but it just doesn't cut it for vision work, and qwen models do.

Re: Qwen 3.7 Preview

#10

Post by tedivm » Mon, May 18, 2026, 5:38 PM UTC

rspoerri wrote: post #4

I am very interested in seeing new qwen models. Qwen3.6 27b is the first one that can do things and doesnt constantly loose "it's mind" and that can be run on a 3090 with a good context size. But it's sometimes getting into a loop.

I've completely replaced GitHub Copilot using Sonnet 3.6 with OpenCode using Qwen3.6 27b, and it's been a great experience.