Missing from the comparison is MiMo V2 Flash (not Pro), which I think could put up a good fight against Step 3.5 Flash.
Pricing is essentially the same:
MiMo V2 Flash: $0.09/M input, $0.29/M output
Step 3.5 Flash: $0.10/M input, $0.30/M output
MiMo has 41 vs 38 for Step on the Artificial Analysis Intelligence Index, but it's 49 vs 52 for Step on their Agentic Index.
skysniper 3 minutes ago [-]
I will try and add it. But I doubt it works well because Mimo V2 Pro is beaten by stepfun even at performance leaderboard (price is not a factor in this leaderboard), so I expect MiMo V2 Flash to perform even worse.
I'm not aware of other AI labs that released base checkpoint for models in this size class. Qwen released some base models for 3.5, but the biggest one is the 35B checkpoint.
thanks for the info. before running the bench i only tried it in arena.ai type of tasks and it was not impressive. i didn't expect it to be that good at agentic tasks
grimm8080 52 minutes ago [-]
Yet when I tried it it did absymal compared to Gemini 2.5 Flash
skysniper 47 minutes ago [-]
what kind of tasks did you try?
grigio 5 minutes ago [-]
i like StepFun 3.5 Flash, a good tradeoff
hadlock 3 hours ago [-]
According to openrouter.ai it looks like StepFun 3.5 Flash is the most popular model at 3.5T tokens, vs GLM 5 Turbo at 2.5T tokens. Claude Sonnet is in 5th place with 1.05T tokens. Which isn't super suprising as StepFun is ~about 5% the price of Sonnet.
It was free for a long time. That usually skews the statistics. It was the same with grok-code-fast1.
MaxikCZ 2 hours ago [-]
Exactly. When I read the headline I thought: "Ofc it is, its free."
skysniper 2 hours ago [-]
I should have clarified I didn't use the free version...
3 hours ago [-]
skysniper 3 hours ago [-]
the real surprising part to me is that, despite being the cheapest model on board, stepfun is often able to score high at pure performance. Other models at the same price range (e.g. kimi) fails to do that.
smallerize 3 hours ago [-]
It looks like Unsloth had trouble generating their dynamic quantized versions of this model, deleted the broken files, then never published an update.
dmazin 2 hours ago [-]
why do half the comments here read like ai trying to boost some sort of scam?
2 hours ago [-]
sunaookami 43 minutes ago [-]
Tried the free version on OpenRouter with pi.dev and it's competent at tool calling and creative writing is "good enough" for me (more "natural Claude-level" and not robotic GPT-slop level) but it makes some grave mistakes (had some Hanzi in the output once and typos in words) so it may be good with "simple" agentic workflows but it's definitely not made for programming nor made for long writing.
skysniper 24 minutes ago [-]
it's actually pretty good at openclaw type of tasks for non technical users: lots of tool calls, some simple programing
skysniper 2 hours ago [-]
another thing from the bench I didn't expect: gemini 3.1 pro is very unreliable at using skills. sometimes it just reads the skill and decide to do nothing, while opus/sonnet 4.6 and gpt 5.4 never have this issue.
skysniper 3 hours ago [-]
I ran 300+ benchmarks across 15 models in OpenClaw and published two separate leaderboards: performance and cost-effectiveness.
The two boards look nothing alike. Top 3 performance: Claude Opus 4.6, GPT-5.4, Claude Sonnet 4.6. Top 3 cost-effectiveness: StepFun 3.5 Flash, Grok 4.1 Fast, MiniMax M2.7.
The most dramatic split: Claude Opus 4.6 is #1 on performance but #14 on cost-effectiveness. StepFun 3.5 Flash is #1 cost-effectiveness, #5 performance.
Other surprises: GLM-5 Turbo, Xiaomi MiMo v2 Pro, and MiniMax M2.7 all outrank Gemini 3.1 Pro on performance.
Rankings use relative ordering only (not raw scores) fed into a grouped Plackett-Luce model with bootstrap CIs. Same principle as Chatbot Arena — absolute scores are noisy, but "A beat B" is reliable. Full methodology: https://app.uniclaw.ai/arena/leaderboard/methodology?via=hn
I built this as part of OpenClaw Arena — submit any task, pick 2-5 models, a judge agent evaluates in a fresh VM. Public benchmarks are free.
vessenes 35 minutes ago [-]
Cheapest just isn't a very useful metric. Can I suggest a Pareto-curve type representation? Cost / request vs ELO would be useful and you have all the data.
skysniper 8 minutes ago [-]
TBH that was my initial thought too, but I found some problem using this approach:
Essentially I'm using the relative rank in each battle to fit a latent strength for each model, and then use a nonlinear function to map the latent strength to Elo just for human readability. The map function is actually arbitrary as long as it's a monotonically increasing function so it preserves the rank. The only reliable result (that is invariant to the choice of the function) is the relative rank of models.
That being said, if I use score/cost as metrics, the rank completely depends on the function I choose, like I can choose a more super-linear function to make high performance model rank higher in score/cost board, or use a more sub-linear function to make low performance model rank higher.
That's why I eventually tried another (the current) approach: let judge give relative rank of models just by looking at cost-effectiveness (consider both performance and cost), and compute the cost-effectiveness leaderboard directly, so the score mapping function does not affect the leaderboard at all.
johndough 2 hours ago [-]
Could you add a column for time or number of tokens? Some models take forever because of their excessive reasoning chains.
skysniper 1 hours ago [-]
both are shown in battle detail page already. Time is shown in Scores table. Number of tokens are shown in Cost details at the bottom of the Scores. (I thought most people just want to see cost in USD so I put token details at the bottom)
hadlock 8 minutes ago [-]
some kind of top-level metric like avg tokens/task would be useful. e.g. yes stepfun is 5% the price of sonnet, but does it use 1x, 10x or 1000x more tokens to accomplish similar tasks/median per task. for example I am willing to eat a 20% quality dive from sonnet if the token use is < 10% more than sonnet. if token use is 1000x then that's something I want to know.
refulgentis 3 hours ago [-]
Please don’t use AI to write comments, it cuts against HN guidelines.
3 hours ago [-]
skysniper 3 hours ago [-]
sorry didn't know that. Here is my hand writing tldr:
gemini is very unreliable at using skills, often just read skills and decide to do nothing.
stepfun leads cost-effectiveness leaderboard.
ranking really depends on tasks, better try your own task.
refulgentis 3 hours ago [-]
It’s too late once it’s happened. I was curious, then when I saw the site looked vibecoded and you’re commenting with AI, I decided to stop trying to reason through the discrepancies between what was claimed and what’s on the site (ex. 300 battles vs. only a handful in site data).
rat9988 2 hours ago [-]
Too late for what? For you? maybe. There are many others that are okay with it and it doesn't disminish the quality of the work. Props to the author.
refulgentis 2 hours ago [-]
> Too late for what? For you? maybe.
Maybe? :)
> There are many others that are okay with it
Correct.
> and it doesn't disminish the quality of the work.
It does affect incoming people hearing about the work.
I applaud your instinct to defend someone who put in effort. It's one of the most important things we can do.
Another important thing we can do for them is be honest about our own reactions. It's not sunshine and rainbows on its face, but, it is generous. Mostly because A) it takes time B) other people might see red and harangue you for it.
skysniper 3 hours ago [-]
all 300+ battle data are available at https://app.uniclaw.ai/arena/battles, every single battle is shown with raw conversional history, produced files, judge's verdict and final scores
refulgentis 2 hours ago [-]
Thanks! Is the judge an LLM? There's lot of references to "just like LMArena", but LMArena is human evaluated?
skysniper 2 hours ago [-]
> Is the judge an LLM?
Yes, judge is one of opus 4.6, gpt 5.4, gemini 3.1 pro (submitter can choose). Self judge (judge model is also one of the participants) is excluded when computing ranking.
> There's lot of references to "just like LMArena", but LMArena is human evaluated?
Yeah LMArena is human evaluated, but here i found it not practical to gather enough human evaluation data because the effort it take to compare the result is much higher:
- for code, judge needs to read through it to check code quality, and actually run it to see the output
- when producing a webpage or a document, judge needs to check the content and layout visually
- when anything goes wrong, judge needs to read the execution log to see whether partial credit shall be granted
if you look at the cost details of each battle (available at the bottom of battle detail page), judge typically cost more than any participant model.
if we evaluate with human, i would say each evaluation can easily take ~5-10 min
refulgentis 2 hours ago [-]
Fair enough, yeah, agent evals are hard especially across N models :/
Thanks for replying btw, didn't mean any disrespect, good on you for not getting aggro about feedback
skysniper 1 hours ago [-]
I appreciate honest feedback, best way to learn :)
3 hours ago [-]
citizenpaul 1 hours ago [-]
>Other surprises: GLM-5 Turbo, Xiaomi MiMo v2 Pro, and MiniMax M2.7 all outrank Gemini 3.1 Pro on performance
This has also been my subjective experience But has also been objective in terms of cost.
Rendered at 19:37:30 GMT+0000 (Coordinated Universal Time) with Vercel.
Pricing is essentially the same: MiMo V2 Flash: $0.09/M input, $0.29/M output Step 3.5 Flash: $0.10/M input, $0.30/M output
MiMo has 41 vs 38 for Step on the Artificial Analysis Intelligence Index, but it's 49 vs 52 for Step on their Agentic Index.
If you haven’t heard of it yet there’s some good discussion here: https://news.ycombinator.com/item?id=47069179
- https://huggingface.co/stepfun-ai/Step-3.5-Flash-Base
- https://huggingface.co/stepfun-ai/Step-3.5-Flash-Base-Midtra...
I'm not aware of other AI labs that released base checkpoint for models in this size class. Qwen released some base models for 3.5, but the biggest one is the 35B checkpoint.
They also released the entire training pipeline:
- https://huggingface.co/datasets/stepfun-ai/Step-3.5-Flash-SF...
- https://github.com/stepfun-ai/SteptronOss
https://openrouter.ai/apps?url=https%3A%2F%2Fopenclaw.ai%2F
It was free for a long time. That usually skews the statistics. It was the same with grok-code-fast1.
The two boards look nothing alike. Top 3 performance: Claude Opus 4.6, GPT-5.4, Claude Sonnet 4.6. Top 3 cost-effectiveness: StepFun 3.5 Flash, Grok 4.1 Fast, MiniMax M2.7.
The most dramatic split: Claude Opus 4.6 is #1 on performance but #14 on cost-effectiveness. StepFun 3.5 Flash is #1 cost-effectiveness, #5 performance.
Other surprises: GLM-5 Turbo, Xiaomi MiMo v2 Pro, and MiniMax M2.7 all outrank Gemini 3.1 Pro on performance.
Rankings use relative ordering only (not raw scores) fed into a grouped Plackett-Luce model with bootstrap CIs. Same principle as Chatbot Arena — absolute scores are noisy, but "A beat B" is reliable. Full methodology: https://app.uniclaw.ai/arena/leaderboard/methodology?via=hn
I built this as part of OpenClaw Arena — submit any task, pick 2-5 models, a judge agent evaluates in a fresh VM. Public benchmarks are free.
Essentially I'm using the relative rank in each battle to fit a latent strength for each model, and then use a nonlinear function to map the latent strength to Elo just for human readability. The map function is actually arbitrary as long as it's a monotonically increasing function so it preserves the rank. The only reliable result (that is invariant to the choice of the function) is the relative rank of models.
That being said, if I use score/cost as metrics, the rank completely depends on the function I choose, like I can choose a more super-linear function to make high performance model rank higher in score/cost board, or use a more sub-linear function to make low performance model rank higher.
That's why I eventually tried another (the current) approach: let judge give relative rank of models just by looking at cost-effectiveness (consider both performance and cost), and compute the cost-effectiveness leaderboard directly, so the score mapping function does not affect the leaderboard at all.
gemini is very unreliable at using skills, often just read skills and decide to do nothing.
stepfun leads cost-effectiveness leaderboard.
ranking really depends on tasks, better try your own task.
Maybe? :)
> There are many others that are okay with it
Correct.
> and it doesn't disminish the quality of the work.
It does affect incoming people hearing about the work.
I applaud your instinct to defend someone who put in effort. It's one of the most important things we can do.
Another important thing we can do for them is be honest about our own reactions. It's not sunshine and rainbows on its face, but, it is generous. Mostly because A) it takes time B) other people might see red and harangue you for it.
Yes, judge is one of opus 4.6, gpt 5.4, gemini 3.1 pro (submitter can choose). Self judge (judge model is also one of the participants) is excluded when computing ranking.
> There's lot of references to "just like LMArena", but LMArena is human evaluated?
Yeah LMArena is human evaluated, but here i found it not practical to gather enough human evaluation data because the effort it take to compare the result is much higher:
- for code, judge needs to read through it to check code quality, and actually run it to see the output
- when producing a webpage or a document, judge needs to check the content and layout visually
- when anything goes wrong, judge needs to read the execution log to see whether partial credit shall be granted
if you look at the cost details of each battle (available at the bottom of battle detail page), judge typically cost more than any participant model.
if we evaluate with human, i would say each evaluation can easily take ~5-10 min
Thanks for replying btw, didn't mean any disrespect, good on you for not getting aggro about feedback
This has also been my subjective experience But has also been objective in terms of cost.