Comment by dannyw
2 days ago
Trading blows in some benchmarks is a bit exaggerated. If you try the model, `xhigh` is basically feels like the `max` mode (i.e. massive thinker and extremely presistent), and the amount of world knowledge and intent understanding is nowhere close to an Opus class model even from 6mo ago.
It's still very useful, and it'll probably displace a good bit of API spend; but it's not really "trading blows with SOTA from just a half year ago". A bit overblown on the Anthropic/OpenAI has missed the window I reckon.
Also, on the open weight frontier side, Kimi K3 is pretty expensive, and Deepseek V4 Pro/Flash is getting a little less juicy with price increases.
Why would anyone rely on the world knowledge built into a model when the harness can just let it search for current information?
Intent understanding is a big point for sure, but world knowledge I'm not sure I see a use case for it.
That's a reasonable point, so let me qualify: at a glance, it seems Qwen 3.8 27B can trade blows with Opus 4.6 on coding tasks, where trading blows doesn't necessarily mean it's a clear winner or even an equal, but does mean it'll at least hold its own and land a punch or two.
(Which I still think is bonkers, FWIW.)
Opus 4.6 is an especially interesting comparison point, I think, because it was a step change; IMO it's when LLMs became serviceable for coding. Yeah, pre-4.6 models did output code, and that code often superficially worked; and bringing it up to production standards still generally meant rewriting it entirely. Opus 4.6 is when that changed.
From my early tests, it's looking like the public benchmarks are not misleading, and Qwen 3.8 somehow got there too, by and large. I've got a few personal tests. One is a mid-complexity one-shot, purposefully underspecified. Beyond a few minor bugs that it could easily fix once pointed out, Qwen 3.8 largely aced it. There are a bunch of things I'd improve, but that was true of Opus 4.6's output too, and by and large the code is clean and well structured. Also worth noting that I'm running Qwen 3.8 fairly aggressively quantized to fit in VRAM; I'd expect tighter results still from the full weights.
Another test I ran is a variant of a common puzzle with an additional structural constraint that makes the usual solution inapplicable, so the model has to actively turn away from the well-known solution and construct a new one that takes the constraint into account. I've never seen a home model pass that test. Kimi K3 passes it, GLM 5.2 passes it (painstakingly). Qwen 3.8 struggles a lot... but does arrive at a correct solution. First time I see a home model do so.
I haven't yet tested it on long multi-turn scenarios. In my experience, that's where pocket models are weakest against heavyweight ones, especially when quantized.
That aside, it does seem like Qwen 3.8 can, in fact, trade blows with Opus 4.6. I don't know yet if it could replace it, and my money would be on no, but I may well be wrong about that considering how weirdly capable it is.
Interestingly, Qwen 3.8's MTP layer is uncannily accurate too. It still gave me good results up to 6 to 8 predicted tokens, which boosts its speed so much it's competitive with Qwen 3.6 MoE. So that's another bizarrely impressive thing about it.
And given all of the above, I do think that the trillions of dollars invested into OpenAI and Anthropic are becoming harder and harder to justify.
Lacking world knowledge is fine for me, I rarely want to rely on the model’s training anyway when there are plenty of great search options to integrate with.