Qwen 3.8 • Hacker News Board

Re: Qwen 3.8

#31

Post by fhub » Sun, Jul 19, 2026, 9:26 AM UTC

gardnr wrote: post #6

adrian_b wrote: post #4

I assume that this announcement has been prompted by that of Moonshot AI, which has just announced a 2.8T parameter open-weights LLM, Kimi K3, to be published on Huggingface by 27 July. Now the response of Alibaba is that they will also publish soon a big open weights LLM, the 2.4T parameter Qwen 3.8. I wonder if Alibaba has always planned to make this big LLM open weights, or they have chosen to do this now, to bett…

It's hard to say what their motivation is. The Chinese firms seem to be working hard to commoditize intelligence which may be the most effective way to debase American frontier labs. And yeah: it also happens to be really good for humanity.

China is watching world sentiment shifting away from USA. Doing many small things that show both strength and openness is surely very intentional.

Re: Qwen 3.8

#32

Post by mft_ » Sun, Jul 19, 2026, 9:26 AM UTC

nsbk wrote: post #19

Bring it on! Hoping that they release smaller sizes of Qwen3.8. I use the 35B MoE and 27B dense models locally and most of the time I don’t need to reach out to Claude. Extremely useful specially when requests include sensitive and/or personal data

I think everyone is hoping this!

It would be great if they'd release an MoE model somewhere between the 35B size of 3.6 and the 122B version of 3.5 - it could be a great balance of speed and ability for people with reasonably powerful but not insane home computers.

Re: Qwen 3.8

#33

Post by reckless » Sun, Jul 19, 2026, 9:30 AM UTC

hodgehog11 wrote: post #14

The "second only to Fable 5" comment is pretty telling here. I remember early on when a lot of naysayers were saying that Fable was barely an improvement on Opus. Like it or not, Anthropic have a genuine moat right now with that model, provided they continue to allow people to use it. It will be genuinely exciting when an open model is able to beat it.

5.6-sol would be a better comparison given it's general availability and usage allowances

Re: Qwen 3.8

#34

Post by SwellJoe » Sun, Jul 19, 2026, 9:30 AM UTC

Qwen is the most censored of the Chinese models in my testing, which makes me wonder in what other ways it is compromised. Open weights doesn't really reveal what's in there. And, in my tests, existing Qwen models are not at the pareto frontier of any metric; DeepSeek V4 Pro is better, faster, and much cheaper than Qwen 3.7 Max. (DeepSeek is also among the least censored of the Chinese models.)

I guess we'll see if the "second only to Fable" hype pans out. In my limited experience with Kimi K3 (I signed up for a month of the $19 plan) it's slower and chews a lot more, so ends up being pretty expensive; one little feature burned through almost the entirety of my five hour limit. The $20 GPT plan is a lot more useful and includes 5.6 Sol, which is fast and token-efficient enough to be quite usable even with the small plan.

Re: Qwen 3.8

#35

Post by dannyw » Sun, Jul 19, 2026, 9:30 AM UTC

revolvingthrow wrote: post #16

The few tests I ran were by no means comprehensive, but while kimi felt like the real deal qwen seems a bit of a benchmark princess.

Qwen3.6 is still the best agentic open weight LLM around 30b params (Gemma isn’t very good at agentic execution).

I also find the model is a lot more predictable and less “glitchy” when made to think in Chinese. You can do this in the system prompt.

Re: Qwen 3.8

#36

Post by walrus01 » Sun, Jul 19, 2026, 9:30 AM UTC

adrian_b wrote: post #4

I assume that this announcement has been prompted by that of Moonshot AI, which has just announced a 2.8T parameter open-weights LLM, Kimi K3, to be published on Huggingface by 27 July. Now the response of Alibaba is that they will also publish soon a big open weights LLM, the 2.4T parameter Qwen 3.8. I wonder if Alibaba has always planned to make this big LLM open weights, or they have chosen to do this now, to bett…

GLM5.2 being released is also likely a factor

Re: Qwen 3.8

#37

Post by KronisLV » Sun, Jul 19, 2026, 9:32 AM UTC

adrian_b wrote: post #4

I assume that this announcement has been prompted by that of Moonshot AI, which has just announced a 2.8T parameter open-weights LLM, Kimi K3, to be published on Huggingface by 27 July. Now the response of Alibaba is that they will also publish soon a big open weights LLM, the 2.4T parameter Qwen 3.8. I wonder if Alibaba has always planned to make this big LLM open weights, or they have chosen to do this now, to bett…

I just hope that they’ll soon also have like 35B or 80B (like the older Qwen3 Next or thereabout) MoE models that can be run locally.

Like, throw us a bone, we all know we need SOTA for lots of dev work anyways, but at least some tasks can be local.

Re: Qwen 3.8

#38

Post by dannyw » Sun, Jul 19, 2026, 9:32 AM UTC

khurs wrote: post #12

Go China, screw America* *within the scope of open models only

I like my Apache 2.0 licensed Gemma, and NVIDIA’s Nemotrons are decent bases for finetuning or continued pretraining, esp thanks to good documentation and tooling.

Oh, and Mira’s thinking machines lab dropped Inkling, a ~1T open weight model too.

This isn’t US vs China. This is open vs closed.

Re: Qwen 3.8

#39

Post by ferrouswheel » Sun, Jul 19, 2026, 9:33 AM UTC

hodgehog11 wrote: post #14

The "second only to Fable 5" comment is pretty telling here. I remember early on when a lot of naysayers were saying that Fable was barely an improvement on Opus. Like it or not, Anthropic have a genuine moat right now with that model, provided they continue to allow people to use it. It will be genuinely exciting when an open model is able to beat it.

I dunno, I find Fable slops alot. Sol is my workhorse. Fable can be creative but isn't very good at doing work reliably (or without endlessly burning tokens).

Re: Qwen 3.8

#40

Post by InsideOutSanta » Sun, Jul 19, 2026, 9:33 AM UTC

hodgehog11 wrote: post #14

The "second only to Fable 5" comment is pretty telling here. I remember early on when a lot of naysayers were saying that Fable was barely an improvement on Opus. Like it or not, Anthropic have a genuine moat right now with that model, provided they continue to allow people to use it. It will be genuinely exciting when an open model is able to beat it.

I wouldn't call it a moat, but I would call it a noticeably better model. Subjectively, for my own work, I would rate the top models Fable > K3 > Sol.

But it's not like Fable is so substantially better than the other two that I would be seriously impacted if I didn't have access to it anymore. All three are amazing models, and of the three, Fable is the only one that regularly triggers refusals.