I'd guess it's mostly just an artifact of the strength of their Agentic RL environments for those specific domains. If they had the same kind of access to massive amounts of institutional software knowledge, compute infrastructure, and hundreds of millions of users dropping them endless highly technical agent traces, there may have been similar disparities on the agentic benchmarks in other fields.

It's possible that they may have generally capped out for some reason... but, I'd rather be bullish on a new American AI lab making it to first place among western open-weight models that can run on a common consumer device. (Closest in size that's better across that range of benchmarks is Inkling-Small I think, at 276B params, unless I'm forgetting someone.)

Training a mid-size model like this can take anywhere between $500k to like $5m. Liquid is spending money on GPU-hours (not sure if you'd call that "energy"). The amortization question is harder to answer without knowing their financials.

The license doesn't really make much sense to me, because if you're already at $10m revenue, you can probably train your own (better, more tailor-suited) model, or at the very least raise money to do so.

I don't agree with that conclusion. I'm looking at the benchmarks page for LFM2.5-24B-A2B, and it appears to get a worse score on SWE-Bench-Verified (by a lot) when compared to Qwen 3.6 35B-A3B. It only clearly beats Qwen 3.6 35B on AutomationBench, ToolSandbox, and LiveClaw.

At least for my use cases, Qwen 3.6 35B-A3B appears better than this new release. An MoE for Qwen 3.8 would be even better than that!

Don't waste time on bonsai unless you have like 16gb or less ram (ram, not vram). Although I'm pretty sure you could still run this model well on 16+8, and qwen3.6 35b-a3b runs well on ~32gb combined ram at q4. Just offload most of the experts to cpu/ram if you have limited vram budget. With the moe models it doesn't hurt performance too much.

But bonsai is basically trash on real agentic coding work. I've tried both bonsai models, all the quants, messed with the hyperparameters, tried to fine tune (didn't succeed), and it just kept looping or failing to finish tasks. Qwen3.6 35b-a3b is my current go to for agentic work for any systems under 24gb vram, although honestly I'm not sure the dense qwen3.8 27b is meaningfully better. The main difference I see is it does a lot more testing attempts and burns a ton of tokens + compactions, so every task it takes 1.2-2x as long as 35b-a3b does. But the quality of the results is comparable.

Same. Asked for some questions and it put out some old knowledge as well.

fast is fast (18tk/s on a Intel Core 7 155H laptop), but quality is something else

How about if you let use with web search, tool calls, and RAG? Does iimprove?

> Qwen 3.8 was slower

Is that Qwen 3.8 27B? Because yes, you'd expect a plain 27B model to be better than a 24B-A2B MoE model all else being equal.

(All else isn't equal though of course - Qwen models are very, very good at coding)

Gemma-4 26BA4B would be a more similar comparison.

That's not surprising. Even Liquid's own documentation states that coding isn't this model's strong suit. Its strength lies in tool calling.

Have you tried Gemma-4 26BA4B? I feel like it is underrated for tasks like that, but maybe I've just been lucky with my use cases.

in my experience it's stronger for terminal use, like chaining bash calls, monitoring background tasks, etc. rather than for straight up coding/swe work. Good with web search and file search rather than just editing software.

Claude code is 60-80 tps, codex varies by model but mostly 40-50 tps.

>How many tokens/s do you need for a coding agent to feel like Claude or codex?

it's kind of a loaded question, this model gets a lot of tok/s but reasoning is forced on; the time-to-answer is really a combination of things like "how much does this thing want to over-reason" and "how much does this thing like tool use?" .

AFAIU prefill speed makes a bigger difference in agentic coding workflows as the harness likely spends more time reading than creating new code. Qwen in particular spends a lot of time reasoning so it does make a larger impact there

I’d say only 65