Hacker news

  • Top
  • New
  • Past
  • Ask
  • Show
  • Jobs

Qwen4-Max (https://qwen.ai)

882 points by ilreb 4 days ago | 485 comments | View on ycombinator

simonw 4 days ago |

Qwen4-Max pelicans at four reasoning efforts (none, low, high, xhigh): https://tools.simonwillison.net/markdown-svg-renderer?url=ht...

For comparison, here's one I got from Qwen3.8 Max (at default reasoning effort): https://simonwillison.net/2026/Jul/20/afraid-of-chinese-mode...

This inspired me to re-run pelicans through 3.8 Max - but OpenRouter has qwen3.8-max-prime and qwen3.8-max-0902 - prime is twice the price and faster - so I went with the 0902 one.

Turns out Qwen4-Max has a none setting, but 3.8 Max didn't allow disabling reasoning - you had to pick from these five levels instead: minimal, low, medium, high, xhigh.

Here are those 3.8 Max 0902 pelicans: https://tools.simonwillison.net/markdown-svg-renderer?url=ht... - they're mostly not as good as 4, but I prefer the 3.8 xhigh one.

maplet 4 days ago |

It scores a 43 on Artificial Analysis with an average cost of $3.27 [1]. For reference, GPT-6.1 Sol (medium) gets a 46 with an average of $0.36.

[1]: https://artificialanalysis.ai/models/qwen4-max

seizethecheese 4 days ago |

I hope they are able to serve 4 faster. Qwen 3.8 is a great model but I’ve regretted including it in the battle system for opencode [1] we have been building because it scores so low.

And it’s a shame because the only reason it scores so low is that it’s slooooow. (Like 30 tokens per second vs opus near 80.) If you exclude voting, Qwen 3.8 does well.

(We allow early voting contrary to Arena.ai because I think it makes sense for agents, where it doesn’t for chat. This means our data takes speed into account.)

Also, it would be ideal if they served this with third parties. Part of the slowness with 3.8 is just not a lot of providers. Their open source version is much faster, but we couldn’t include it because it doesn’t handle images (another own goal by Alibaba).

[1]: https://agentduel.com/leaderboard

hauntingseaweed 4 days ago |

Anthropic deserves to go bankrupt for trying to ban open-weight/open-source competition. I hope their IPO flops.

dan_thinks 4 days ago |

This is exciting. The continued focus on efficiency alongside quality is encouraging - it makes me hopeful to see what smaller models will become capable of.

Looking forward to the "Flash" and 27B versions.

pu_pe 4 days ago |

I think Alibaba is not aiming for the frontier on this class. Their previous Qwen3-Max was released in September, right around when Opus 5.5 was coming out, and it was nowhere close to it as well. However, one or two tiers below (Flash, 27B) they really shine and are SOTA or close to it.

DeepSeek has a similar problem, I feel like their full model is hardly even worth using, while the Flash tier is absolutely amazing.

spwa4 4 days ago |

So what is happening is that models are broadening difficult-to-reason-to skills on benchmarks. If you take the traditional benchmarks (ie. DeepSWE, Terminal-Bench slightly older version, GDPVal, Automation Bench, ...) this model is between Sonnet and Opus 5.5, VERY close to the frontier.

What has happened to artificial analysis is that they've begin to focus a lot on very broad tasks like legal reasoning, using very difficult to work with scientific software packages, correctly analyzing biology experiments (without or only tiny amounts of explanation, count cells, BUT leave it to the model to figure that cell that aren't round, or have a distorted cell membrane (ie. have burst) are dead), very difficult physics reasoning ("2 electrical wires are close together with power flowing in the same direction. What happens?", and you get points deducted if you don't realize the problem with high voltage AC power, where the wires will shake themselves loose if the phase doesn't match perfectly, meaning the wires need to be EXACTLY the same length or ...). If I'm being impolite I'd say the frontier models are trying to make sure they'll pass the "gotcha" interview questions for many jobs.

The 3 differences:

Qwen 4 is a huge jump from Qwen 3.8 in coding

Qwen 4 is a big jump in using it for agentic workflows

Qwen 4 also performs much better on avoiding hallucination

If the difference between Qwen 4 Max and Qwen 4 flash is similar to 3.8 and 3.8 flash next (ie. about 10%), then 3.8 flash next may just barely beat Opus 5.4 at coding.

nelaggy 4 days ago |

is the open weights lag widening? these open models don't seem to be improving as quickly as ant/oai

m00dy 4 days ago |

I bet most of us waiting for 27b model and want to see how good it will be.

WASDx 4 days ago |

Warning: The website downloads more than a gigabyte of videos before you even start watching anything.

jhonof 4 days ago |

Any info on when the 27B or FN Qwen 4 models are coming out?

xpct 4 days ago |

I was waiting for whether there'll be Chinese model drops before the Anthropic IPO. Still a few days left, wonder if there's more coming.

anana_ 4 days ago |

Artificial Analysis score doesn't look too hot. Doesn't even surpass Kimi 3.1

andy_ppp 4 days ago |

Quite a few saying that they tried earlier snapshots of this model and are finding it does not live up to expectations. Firstly do other agree and also why would the model get worse - could they be doing something weird to the model that is making less aligned at the end of training, potentially for political reasons?

husseinhb 4 days ago |

I was waiting GLM from z.ai and it still not releasing, almost all the models releases their new version of models and they still on 5.3 who is very weak in complex tasks and do half of the work and forgets to make migrations and like that issues. They sell us the yearly plan and stopped improving. Because of that i am just using in coding opus 5.5 and in chat fable they are the best models from experience.

drnick1 4 days ago |

Are these (hosted) Chinese models still accepting arbitrary cybersecurity and reverse engineering tasks? I am increasingly frustrated with Claude's safeguards (and Codex, to a much lesser degree). It is flat out refusing to look at game binaries and anti-cheats for the sake of learning and devising Linux compatibility workarounds. It seems like in the near future people who want to do this type of work will need to buy a couple of DGX Sparks or retired datacenter GPUs and run abliterated open models on their own hardware.

libpcap 4 days ago |

Does it provide an objective response when asked, "What happened in China on June 4, 1989"?

chriddyp 4 days ago |

In our short horizon data analysts exam, it scored a few points worse than Qwen 3.8 27B from August (which is open weights) and cost a bit more. There’s about 10 models that we’ve tested that have scored both better and cheaper. Too bad it’s closed, Qwen 3.8 27B really punched above its weight and was the only open model I’ve tested that could run on desktop hardware and perform very well: https://plotly.com/blog/qwen4-max-plotly-data-analytics-benc...

Luke2642 4 days ago |

I wonder if they could benchmax Program Bench, it's one of the few benchmarks that looks almost exactly like a good fine tuning curriculum which might help general coding ability and agential work across the board?

I suppose it could make actually worse at taking instructions and redirection mid-task, but that could be easy to work around too. Just hold back a bit of the spec and drip feed it in later?

I don't mean cheat on it, I just mean optimize for it harder!

mlmonkey 4 days ago |

Interesting that it's only at a reasoning effort of XHigh does the pelican's left leg actually go behind the bicycle. At lower reasoning levels the 3D aspect of the figure is not correct.

akulbe 4 days ago |

I subscribed to the Lite plan, but I'm not finding where I can get/generate an API key, to be able to use it with something like Hermes Agent?

dbg31415 4 days ago |

Site seems to have been slashdotted.

thatxliner 4 days ago |

Where model weights?

grendelt 4 days ago |

It's great fun asking it questions like "who is the Dalai Lama?", "where is TSMC headquartered?", "why do people post Winnie the Pooh in reference to Xi?", "who controls the South China Sea?", "why was Chiang Kai Shek fighting Mao?", "why were students in Tiananmen Square?"

It'll gush about how to make moon cakes and where the tallest and longest bridges are and how to run qwen on a raspberry pi - but it can't seem to manage some basic Wikipedia questions.

syntaxing 4 days ago |

Every time a major open weight is released, artificial analysis updates to a new one (v5) now. Exaggeration but some of this feels very apt (its also hilarious) https://www.reddit.com/r/LocalLLM/comments/1xkhha2/mr_incred...

expectfun 4 days ago |

Funny how other flagship model releases feel like "Great, let's see how far they pushed frontier" and Qwen Max releases feel like "oh nevermind, what about the 27B version?"

There's plenty of competitors in the league where Max is, but few companies can improve the 27B level capabilities.

HarHarVeryFunny 4 days ago |

Did Qwen always adopt the AI butthole icon convention, or is this new?

undefined 4 days ago |

undefined

trilogic 4 days ago |

Cybersecurity not mentioned. Can´t wait to test bench the open weights in cybersecurity. If is not performing well by design on it, I imagine someone have to add some additional layers XD.

The rest looks awesome.

nxtfari 4 days ago |

Posting a loss curve is an… interesting marketing approach by Qwen? I don’t know what that’s supposed to inform me about the model.

Onavo 4 days ago |

The good thing is that its biotech has improved (somewhat).

xiaoyu2006 4 days ago |

Disappointing performance for a 2.4T model in the end of 2026.

nixosbestos 4 days ago |

Really would love if these publications have a huge, bright red bolded section "How to use this right now". Is it on openrouter? Is Qwen hosting it already? Maybe I'm just ignorant?

EDIT: oh, I guess this isn't open weights, so presumably I'll have to get it through Qwen or whoever they might partner with (Bedrock?).

Ikersl 4 days ago |

i think its good you can run small models easily, even if they are quite dumb

peter_d_sherman 3 days ago |

[dead]

undefined 3 days ago |

undefined

sodowork 4 days ago |

[flagged]

zhuanxyz 4 days ago |

[flagged]

senko 4 days ago |

[dead]

mipo256 4 days ago |

[flagged]

ankur_misra 4 days ago |

[flagged]

cricketfanatic 4 days ago |

[dead]

Vivek-KY 4 days ago |

[flagged]

jacksonastone 4 days ago |

Anyone else worried these models have poison pills embedded in the weights? (Past examples: https://cohere.com/blog/concerning-behaviour-popular-open-we...)

small_model 4 days ago |

Only 43 on AA, no where near the frontier now, looks like distillation blocking has halted their progress.

bbor 4 days ago |

[flagged]