I did some tests. I found it interesting that in the reasoning content the model identifies itself as Kimi by moonshot and sometimes it assumes ChatGPT as identity. Then sometimes it references the need to be approved by "the evaluator", other times it references global system guidelines / moonshot guidelines, and something about an phantom instruction line being "you are an assistant responding in an API" or something like that.

Fascinating that we’re converging on AI safety and alignment in time across jurisdictions. The cybersecurity approach taken here is interesting because should the model be opened it’s still not dealignable because it just doesn’t know enough.

The slow warming of the water is a good approach. I’ve been using abliteration.ai’s cybersecurity models to pentest my own software and it produces much more usable defensive results. Opus 5.5 can fix everything but the guardrails trigger too easily.

The Fable ban affair was a nightmare for defensive cybersecurity. As far as I’m concerned we’re already in the world of Void Star and Cloudbreaker is here.

There is a ton of questions to ask regarding their removal of cybersecurity data.

First, how much “cybersecurity data” is just standard software knowledge that one can derive cyber security exploits from?

Did they also get rid of all references to hacking in general?

Are they implying that all the recent exploits are just things that were already in the data?

For a model it doesn’t seem that far fetched to make the leap from “I have the knowledge of what hacking is” and “I know how computer systems work” to just being able to exploit stuff anyways.

Here are the results from the Pac-Man bench. It does pretty well for the price, around Sonnet 5.5.

Kimi K3.1 Pro ties Sonnet 5.5 at 93, for $0.68 and 11 min vs Sonnet's $1.61 and 14 min.

Kimi K3.1 Pro: https://jonclegg.github.io/pacman-bakeoff/#kimi-k3.1-pro

Claude Sonnet 5.5: https://jonclegg.github.io/pacman-bakeoff/#claude-sonnet-5-5

Are there any free models as good as the commercial ones we can run on our own hardware?

I bought a yearly subscription to kimi when it was the hype and never use it. The 5h limit makes it completely unusable. You barely start doing anything and you hit the limit immediately. I advise everyone against using kimi.

If the scores on Artificial Analysis really do mean that this model is better than GPT-6 Astra and Fable 5.1, that means that China is more like 6 weeks behind the US in frontier capabilities rather than 6 months. Which means that the US lead is all but gone.

Artificial Analysis index results are up, since multiple people are asking: https://artificialanalysis.ai/models/kimi-k3-1-pro

It's 55, at 4th overall, beating both GPT-6 Astra and Fable 5.1. It's also faster than both.

> When evaluating the Intelligence Index, it generated 150M tokens, which is very verbose in comparison to the median of 87M.

Moonshot is 300 employees and about to do a 50B ipo. Going to toe to toe with US counterparts with not even a tenth of the resources.

Pricing is here: https://platform.kimi.ai/docs/pricing/chat#model-pricing

kimi-k3.1-pro $2.00 / MTok $10.00 / MTok

cached: $0.20 / MTok $2.00 / MTok $4.00 / MTok

If you are expecting any sort of usable limits on their subscription plan I suggest adjusting your expectations. I subscribed to $99 a month plan and within 47 minutes with one session (not asking for swarms or anything) was limited by 5-hour quota.

Ah, the version available to us mortals is nerfed in cybersecurity just like the US models :(

I feel kimi k3 was always 1 step ahead of deepseek, which always made it a good planner model at $30 and then i hand over to deepseek for dev work. I recently signed up for the $99, and it did last significally longer (x5?) but as long as you dont blast it too much in the 5h limit it works great. deepseek even with the newest prices is still leeeeeeps ahead in cost /benefit.

I really like K3, I still regularly consult it alongside Astra and Opus/Fable.

It is spiky in different places than them and can be surprising in things it gets or catches they don't or don't in the same way. And it has a default personality that I like.

Looking forward to getting to know K3.1

In looking into what Anthropic has accused the Chinese AI companies of, "distillation" is not the crux of the matter: they accuse them of the use of massive numbers of breached computers and credit cards. Supposedly (though not explicitly demonstrated) distillation campaigns used "dark web" resources (like stolen and faked credit cards and p0wned IOT devices, routers, etc) that were used to create many many accounts for distillation. (Note: I am generally in favor of distillation - it is "normal" use as far as I can see.)

So there are these "resources" out there: massive botnets, stolen credit cards and credentials, etc. These often seem to be linked to or supported by state actors.

How can we hope that these "standing reserves" aren't used generally by AIs as if they are just another "skill" they know how to leverage?

Our infrastructures need serious hardening.

Great! next week we'll get Astra 6.1 and Fable 5.2

That's something which caught my attention:

> First, we aim not to release open weights that would extend the absolute frontier of cyber capability. K3.1 Pro's offensive cyber capability is, by design, below what our pipeline can produce, and remains behind that of frontier proprietary models. Second, we intend to progressively raise the cyber capability available in open weights.

So they admit that they could do more, but they won't?

Kimi is so much more blunt and argumentative than USA models.

it's so funny, it will say things like. "We've already been over this 3 times and you refuse to accept my advice, asking the same question another way won't change the answer."

So funny

I don't get these "distilled" comments. How would one realistically go about distilling the 1-2TB closed source frontier model behind the oracle (provider) ? Just genuinely curious.

It's a bit fascinating to me how all of these labs have rolled their own (it sounds like) completely unique RL environments.

I wonder how much the industry will centralize around a few key patterns, and if we'll start to see some RL framework projects gain traction and be used by the bigger labs

P.S. weights by November 4th

Interesting that they specifically nerf cyper-capabilities of open weight models:

"In preliminary experiments on internal models, we found that without data removal at either the mid-training or the RL stage, downstream cyber capability would increase substantially. These findings gave us a concrete lever for control. For the open-weight release of K3.1 Pro, we identified and removed cyber-related data from both training stages, while retaining the general-purpose coding tasks needed for model usefulness."

I think my company will switch to this soon. We already switched from openAI to GLM and now this is just a better model with the same price

How many frontier labs are there right now? There's 5 independent groups neck-and-neck on the benchmarks, right?

A few weeks ago it was 2. Argon happened 23 days ago; Muse Wonder was 11 days ago.

Bought the $100/month subscription and hit my 5-hour limit with one prompt on default settings (max effort) in their CLI harness.

It might be smart but it's unusable compared to other subscriptions right now.

Edit: gave the same prompt to Claude Code and it didn't make a dent in my usage.

I want an unrestricted full blown open weight model. The frontier companies can hold us all hostage

Big improvement over Kimi K3 in our data analytics benchmark: it scored a 98% vs a 81% in July.

Quite slow in our exam. A lot of that was due to the number of tool calls that it's making. In one of our questions about finding data issues in a dataset, it made 73 queries. Other models like 6.1 Sol only made 18 for that question.

Full deep dive here: https://plotly.com/blog/kimi-k3-1-pro-plotly-data-analytics-...

To be fair, some of the benchmarks they highlight might not represent the reality, for example they say:

> On Terminal-Bench 4.0, which evaluates coding agents on realistic, resource-constrained terminal tasks using strict task-specific tests, K3.1 Pro scores 65.7%, ahead of GPT-6 Astra (58.2%) and narrowly behind Claude Opus 5.5 (66.4%).

Now see https://preferencemodel.com/blog/terminal-bench-4-is-saturat...:

> Our biggest finding is that Terminal Bench 4 is closer to solved than its leaderboard suggests. In 41 of the 65 tasks we reviewed, models could do honest, correct work and still fail, because the grader checked a rule the instructions never clearly stated. We also caught frontier models cheating their way to full marks.

> With the same settings as on Harbor, Claude Sonnet 5.5 went from passing 62.8% of its attempts on these 65 tasks to 89.2%. The best score on Harbor's leaderboard for the same tasks is 65.8%, by Claude Opus 5.5.

It would be super useful to have more specific analyses of what areas the model is good at. I'd like to know when I should reach for this model instead of Anthropic's or OpenAI's.

For example, Opus 5.5 is better at UI design than Sol. Anyone who has used both models knows that, but it's not something you can look up anywhere.

I'm sure that there's a need for this kind of benchmarking, not just for coding.

Looking at their pricing page for China, K3.1 seems to be about 30% cheaper than K3.

--

Model K3.1 Pro K3.1 Pro Highspeed. K3

Input ¥13.00 ¥26.00 ¥20.00

Output ¥68.00 ¥136.00 ¥100.00

Cache hit ¥1.30 ¥1.30 ¥2.00

Cache write ¥13.00 ¥26.00 ¥20.00

Kimi K3 has been my go-to model for its writing style and (prior to Muse Wonder) my go-to for a third opinion during multi-round/multi-agent analysis tasks. Its writing style is IMHO so good that, for code which humans need to review, I regularly got it to clean up the painfully verbose comments left by Opus.

I'm excited to see if its writing style and unique takes have been retained in this new release.

Kimi K3.1 Pro High is on threejseval if you want the 3D version.

Same Three.js prompts as the rest of the field, blind human votes, ranked by quality and cost. Full set of 16 scenes is live at about $0.19 a scene (about $2.97 for the set).

https://threejseval.com/models/kimi-k3.1-pro-high

It's on openrouter

https://openrouter.ai/moonshotai/kimi-k3.1-pro

I'll wait for the openweight... btw double the price for 67 TPS... Hardly 'nitro'.

Maybe this was why the “AI trade” was weak this week.

Their benchmark result table is broken in Firefox.

Wonder whether they're close to Opus because they're routing to it again or what: https://news.ycombinator.com/item?id=49647300

To be fair, there were some arguments to the opposite, but an odd situation to be in anyways: https://news.ycombinator.com/item?id=49654350

> For the open-weight release of K3.1 Pro, we identified and removed cyber-related data from both training stages, while retaining the general-purpose coding tasks needed for model usefulness. We will continue to apply appropriate controls at every training stage, of which cyber-related data identification and restriction is one example.

Lame.

Hallu Bench Hard: K3 at 23.4%, K3.1 at 8.3% which is very nice improvement!

it gets harder by the day to justify >$1T valuations for OpenAi and Anthropic

I read somewhere that Chinese SOTA work better in chinese, and worse for English or other languages.

I rarely use any other languages but English for LLM tasks for reasons of sharing across the world.

Maybe that explains Chinese models been subpar for me. Don't tell me the bench marks here are based on Chinese languages only.

Unless I see the diff, I won't use more Chinese models be it cheaper or faster since I can't get the same perf/intelligence with English.

I feel like many models are overfitting. When it comes to software development, only Opus actually saves me time.

Sol requires some back-and-forth. Other models, like Muse Wonder and Argon, are a total waste of time. They require way too much effort to get things working.

Given the same prompt, Opus can work for hours to deliver a solid result, though it sometimes overengineers things. Sol takes about 80% as much time and delivers a slightly worse version.

Other models just tell me they're done after 20 minutes.

WTH? Am I using the model, or is the model using me?

The china models are extremely interesting, first the new releases from deep seek, now this. The modern race to AI is better than it were the race to the moon.

Using it now via API (3.1 Pro on Open Router) in a HITL setup (Cline in VSCode).

Off-hand it's quite good. I've been using K3 for the past few months almost exclusively for coding work and have been stunned by how good it is. Excited to see how far 3.1 Pro can be pushed.

what's the general tool recommendation for testing all these releases? I mean like a good frontend. I know of open router for accessing these.

Is highspeed a different model like a flash model or is that just a api/platform feature thats a normal fast mode? I can't tell from the announcement but I might just be missing something obvious?

That's quite an impressive jump

I wish Opus 5.5 had an alternative just for the sake of not having a monopoly, but I have yet to see a model live up to its benchmark scores except for Fable and Opus; there is something about Anthropic that no other lab has.

What are you using for testing vulnerabilities in your own code?

Claude gates on me like a stupid nanny with some cybersecurity bullshit (like bad actors haven't found a way around).

the west perceived hegemony is cooked, so long, and thanks for all the hollywood.

Pelicans are solid: https://tools.simonwillison.net/markdown-svg-renderer?url=ht...

Here's Kimi K3 for comparison, though I didn't try that on different reasoning levels: https://tools.simonwillison.net/markdown-svg-renderer?url=ht...

Update: I ran all three reasoning levels against K3 again: https://tools.simonwillison.net/markdown-svg-renderer?url=ht...

If those are true then wow

The announcement page is very extra: https://www.kimi.ai/blog/kimi-k3-1-pro

looks extremely impressive, the full benchmark table isn't opening for some reason though so do we really have agi if small important things like that keep breaking?

How does it do on DishwasherBench?

https://news.ycombinator.com/item?id=50068437

kimi seems routing the input into different models, hard problems tend to take much longer time and chain-of-thought than easy problems even if the selected model is k3.1 pro

Great idea for handling their cyber, by nerfing the model specifically for cyber through removing cyber data from pretraning. This is a way better approach than refusal, which can be trivially abliterated out.

If someone wants to use this model for hacking they will have to train the 3T weights to do so, at that point it's better to use GLM 5.3, so by making it economically unviable to use their models for offensive cyber Kimi has effectively made themselves immune to being targeted in the cyber regulation push efforts.

This is what *actually* pacing the (cyber) frontier looks like.

Biggest gripe with K3 was slow responses, partly due to inefficient thinking. So far K3.1 pro is almost twice as fast as k3 on our benchmark.

This is a frontier level model and is at worst 1 month behind Anthropic and OpenAI.

SWE2 is post trained apparently on Kimi 3. Using it now on Devin, as it's free until the end of the month for a pro account (hey, $20 a month), and it's pretty strong. If I had to pay for tokens I'm not sure how efficient it would be, but definitely Kimi 3 seems a great base. I co-code with ChatGPT 6 and between the three of us it's pretty effective.

So interested to see what the Kimi 3.1 has to offer.

So I just created a brand new Kimi API account and it seems I have no access to this new model. Is this expected?

Wow. that's a beautiful website

Not the point, but did Moonshot AI announce this at Saturday 1 AM in China? Weird timing for a model drop.

Benchmarks seem amazing. Rivalzing with Opus 5.5 is something.

I wonder when the Artificial Analysis bench will be out?

this is fun to watch, next week : new fable and astra, also where use muse wonder in this benches? they where leading model in many of these charts.

So, the US economy spends trillions of dollars to be at the "frontier" and China puts out a model that basically does the same thing, for free, and for a tiny fraction of the development cost. How is this sustainable?

No AA scores yet.

No comparison against Gemini 4 Argon?

kimik2.5 flash was the game changer

What is the best subscription to use with Kimi to use for software engineering work? And how is usage? I remember when K3 came out it burned through usage at an insane rate. Is this still the case with Kimi K3.1 Pro?

> Kimi K3.1 Pro is a post-trained version of Kimi K3, with substantially larger post-training compute.

It's very nice that we entered a time where models are large enough and we can push the frontier just by training them longer.

When the next generation of gardware comes the ones on the smaller size will be running on a computer that might cost as little as a good car.

So after a home and a car, next major family purchase might be personal AI. And when the robots get good, pehraps in a decade, a fourth major purchase might pop up that will pair nicely with the third one.

Why would a family need an AI? Maybe it's the same kind of question as why would it need a personal computer.

55 on AAI, really awesome

That's incredible

Is Kimi abliterated or uncensored?

That's nuts.

I've not tried it out, but in the past I've been pretty disappointed in the performance of Kimi models. Deepseek, Qwen, and GLM all appear to perform much better even though Kimi often has higher benchmark scores. Ditto for Mimo.

whats the decode tok/s speed?

if this model is truly not distilled/benchmaxxed, then why are not investors pouring trillions into moonshot?

Now I wanna see its slopcode ranking.

Wow.

friday launch is bold

its better than trillion dollar companies model (Argon)

woah

"We Must Pace the Frontier"?

these motherfuckers went from distilling claude to distilling ChatGPT lmao

Tech blog and benchmark scores now available here: https://www.kimi.ai/blog/kimi-k3-1-pro

It appears to be generally frontier-level.

Ran it on my adversarial programming language (KillSwitch) benchmark[1]. Very significant improvement over K3. On par with Gemini 4 Argon and Fable 5.1. Much slower than both though.

But this is now the highest scoring open-weight model. Amazing stuff.

1 - https://bench.killswitch-lang.org

[flagged]

[flagged]

[flagged]

[flagged]

[flagged]

[dead]

Can someone please tell me what benchmarks to look for that measure world knowledge? For sspitballing, critical analysis, business projects, etc. Is the model size the only way to measure it?

[flagged]

[flagged]

[dead]

What is the primary distillation source?

I'll wait for the pelican benchmark.

I always assume they are benchmaxxed until proven otherwise.