Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Qwen 3.7 Preview

twitter.com

Post reply on HN
Print view View original

51–60 of 100 posts

Re: Qwen 3.7 Preview

#51

Post by storus »

trilogic wrote: post #12

Qwen 3.6 35B (finetuned) is so good that it became standard open weights for everyday use. Is not far at all from proprietary models if you give it tools, skills and agents etc, it can actually finish the job. (Thank you Qwen team, appreciated). Using opensource now we can definitely rely to design from scratch very complicated architecture and build pretty fast the full pack. Wish to see Europe AI unleashed, wake up…

It's 3.7-max; max was never open-weighted before. I don't see any smaller models in that tweet.

Re: Qwen 3.7 Preview

#52

Post by SwellJoe »

svachalek wrote: post #29
mempko wrote: post #26

I love that open weight models are catching up so quickly. Also hilarious how far behind Grok is. I guess demand for Grok must be poor if Anthropic is able to rent resources from xAI.

To play devil's advocate I do feel like Grok has a unique "feel" to it. All the Chinese models feel like GPT or Claude distillations, but Grok has a certain unique way of saying and doing things. But that said, it also feels a year behind the state of the art.

With an Austrian accent, perhaps?

Re: Qwen 3.7 Preview

#54

Post by raffael_de »

I have a tangential question. Provided that it is correct that current proprietary models are offered at below cost-covering rates (I believe this is a consensus if I'm not mistaken¹); what factor (multiplication) would have to be applied approximately to current rates to reach break even?

¹: I think I read this a couple of times but I'm not sure if correct to begin with. Can this be substantiated based on annual financial reporting or other published business metrics by OpenAI, Anthropic et al.?

Re: Qwen 3.7 Preview

#55

Post by jedisct1 »

b3ing wrote: post #41
trilogic wrote: post #12

Qwen 3.6 35B (finetuned) is so good that it became standard open weights for everyday use. Is not far at all from proprietary models if you give it tools, skills and agents etc, it can actually finish the job. (Thank you Qwen team, appreciated). Using opensource now we can definitely rely to design from scratch very complicated architecture and build pretty fast the full pack. Wish to see Europe AI unleashed, wake up…

For coding it’s really bad. Writing is ok, chat is good. It’ll get better but it’s not that close yet

Depends on the language and harness, I guess.

It works really well for me, at least for Python and JavaScript, with swival.dev as a harness.

Re: Qwen 3.7 Preview

#56

Post by kajecounterhack »

jedisct1 wrote: post #55
b3ing wrote: post #41

Earlier quoted context omitted.

For coding it’s really bad. Writing is ok, chat is good. It’ll get better but it’s not that close yet

Depends on the language and harness, I guess. It works really well for me, at least for Python and JavaScript, with swival.dev as a harness.

You should probably disclaimer that you're the author of swival.dev, but nice project :)

Re: Qwen 3.7 Preview

#57

Post by bachmeier »

I'm not much interested in vibe coding (for those who aren't aware that LLMs have other uses). The specific model I've been using with Ollama is hf.co/unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF:UD-Q4_K_XL and it's amazing how fast it is on 64 GB of RAM and i5-13400 CPU. No GPU on this computer. Gemma 4 E4B will think for a couple of minutes vs 3-5 seconds for Qwen. It's hard to believe how much you can do with such limited hardware using their models.

Re: Qwen 3.7 Preview

#58

Post by Computer0 »

genxy wrote: post #37
sleepyeldrazi wrote: post #28

Earlier quoted context omitted.

Finetuning takes little resources, the base model training is the slow and expensive part. Architecturally 3.5 models are identical to their 3.6 counterparts, that is why there is a consensus that those are probably finetunes and not re-trained from scratch, like you will se many people publish their own on huggingface.

Understood, but look at their larger cadence over the years and the breadth of models. They are clearly not all finetunes. Meta for all its billions, doesn't have anything comparable.

competent leadership goes a long way

Re: Qwen 3.7 Preview

#59

Post by Vaskivo »

nyrikki wrote: post #50
rspoerri wrote: post #46

Earlier quoted context omitted.

how do you do 1mio context with qwen3.6 27b, that only supports 256k? and what hardware would you run that on? 2 * 3090 is afaik currently at max 256k context.

You can get all the Qwen 3.x models up to ~1 million tokens using YaRN with llama.cpp.[0] Personally I am using `--no-context-shift` and feeding in context back in on failure at the harness level. I have 2x1080ti + 1xTitanV that have a full 262,144 tokens context on 262,144 tokens with `-sm tensor` at 62.04 t/s which isn't so bad. But I also have a 1x3090 running unsloth/Qwen3.6-27B-MTP-GGUF:UD-Q4_K_XL at 41.89 t/s b…

How can you get it to run at 41 t/s? I also have a single 3090 and even with MTP can't break 20 t/s.

HEre's my setup:

  llama-server
  --port 9999
  --model /MODELS/LLMs/Qwen3.6-27B-UD-Q4_K_XL.gguf
  --ctx-size 128000
  --threads 12
  --flash-attn on
  --device CUDA0
  --jinja
  --gpu-layers 52
  --mmproj /MODELS/LLMs/Qwen3.6-27B-mmproj-F16.gguf
  --cache-type-k q8_0
  --cache-type-v q8_0
  --temp 0.6 --top-k 20 --top-p 0.95 --min-p 0.0 --repeat-penalty 1.0 --presence-penalty 0.0
  --spec-type draft-mtp --spec-draft-n-max 2
(I'm not filling out 100% of the VRAM, as I have other stuff I need it for.)

Re: Qwen 3.7 Preview

#60

Post by bachmeier »

genxy wrote: post #37
sleepyeldrazi wrote: post #28

Earlier quoted context omitted.

Finetuning takes little resources, the base model training is the slow and expensive part. Architecturally 3.5 models are identical to their 3.6 counterparts, that is why there is a consensus that those are probably finetunes and not re-trained from scratch, like you will se many people publish their own on huggingface.

Understood, but look at their larger cadence over the years and the breadth of models. They are clearly not all finetunes. Meta for all its billions, doesn't have anything comparable.

> Meta for all its billions, doesn't have anything comparable.

Maybe nothing released to the public. I don't know that all of their models are public. I think all they really care about is that they aren't relying on one or two cloud providers for a critical piece of their infrastructure.

Post reply on HN