Kimi K2.6
Made by Moonshot, released 20 Apr 2026. Epoch AI lists its access as “Open weights (unrestricted)”. Appears solid at coding among the models listed here.
In short
What it does well
- Coding: Strong (32nd of 77 listed models)
Where it falls short
- Knowledge: Behind the leaders (34th of 43 listed models)
- Maths: Middle (32nd of 43 listed models)
- Writing: Middle (49th of 77 listed models)
What we don't know yet
- Difficult coding tasks: Not enough data
The four ratings
- Top tier
- Strong
- Middle
- Behind the leaders
- Not enough data
Ratings compare only the 77 models listed here, in quarters, best first. “Not enough data” means no result yet. It is never a low score.
- Writing
- MiddleHow often people preferred its answers to writing prompts. 1,433 on Arena's scale · Source for Kimi K2.6 writing score (lmarena)
- Coding
- StrongHow often people preferred its answers to coding questions. 1,489 on Arena's scale · Source for Kimi K2.6 coding score (lmarena)
- Maths
- MiddleHow many very hard maths problems it solved. 57.2% · Source for Kimi K2.6 maths score (epoch)
- Knowledge
- Behind the leadersHow many short factual questions it answered correctly. 34.9% · Source for Kimi K2.6 knowledge score (epoch)
Evidence completeness: Broad evidence, 12 of 13 tests have a result.
Where it is relatively stronger or weaker
This compares each kind of request with the model's own overall position among the 77 listed models, using Arena's votes. It is about performance on different languages, kinds of request and topics, not about political or social bias.
Relatively stronger
No category stands out as clearly stronger than its overall position.
Relatively weaker
No category stands out as clearly weaker than its overall position.
Show every category we can see
| Category | Position | Arena score | Votes | Source |
|---|---|---|---|---|
| Expert-level prompts Kind of request | 32nd of 77 | 1,491 95% range 1,481 to 1,501 | 3,851 | Source for Expert-level prompts |
| Hard prompts Kind of request | 39th of 77 | 1,471 95% range 1,465 to 1,476 | 26,045 | Source for Hard prompts |
| Long prompts Kind of request | 34th of 77 | 1,468 95% range 1,462 to 1,474 | 16,939 | Source for Long prompts |
| Longer conversations Kind of request | 47th of 77 | 1,453 95% range 1,445 to 1,461 | 6,934 | Source for Longer conversations |
| Chinese Language | 26th of 76 | 1,520 95% range 1,508 to 1,533 | 2,313 | Source for Chinese |
| English prompts Language | 39th of 77 | 1,460 95% range 1,455 to 1,466 | 18,154 | Source for English prompts |
| French Language | 43rd of 72 | 1,470 95% range 1,454 to 1,486 | 1,565 | Source for French |
| German Language | 35th of 55 | 1,453 95% range 1,432 to 1,474 | 742 | Source for German |
| Japanese Language | 28th of 54 | 1,442 95% range 1,414 to 1,470 | 499 | Source for Japanese |
| Korean Language | 26th of 54 | 1,427 95% range 1,404 to 1,450 | 734 | Source for Korean |
| Non-English prompts Language | 38th of 77 | 1,446 95% range 1,441 to 1,452 | 21,824 | Source for Non-English prompts |
| Russian Language | 52nd of 77 | 1,446 95% range 1,437 to 1,455 | 4,244 | Source for Russian |
| Spanish Language | 31st of 70 | 1,463 95% range 1,447 to 1,480 | 1,372 | Source for Spanish |
| Entertainment, sport and media Topic | 43rd of 77 | 1,427 95% range 1,420 to 1,435 | 8,282 | Source for Entertainment, sport and media |
| Law and government Topic | 41st of 76 | 1,462 95% range 1,451 to 1,473 | 3,023 | Source for Law and government |
| Medicine and health Topic | 44th of 75 | 1,457 95% range 1,445 to 1,468 | 2,927 | Source for Medicine and health |
| Science Topic | 39th of 77 | 1,468 95% range 1,460 to 1,476 | 6,710 | Source for Science |
| Software and IT Topic | 37th of 77 | 1,484 95% range 1,478 to 1,490 | 16,029 | Source for Software and IT |
Bias: what we can and can't see
Our sources do not measure political lean, stereotyping, flattery (agreeing with you too readily) or safety behaviour, so we cannot say how any model does on those. What we can show is above: where a model is relatively stronger or weaker across languages, kinds of request and topics. Also, people voting in Arena can favour answer style, such as length and formatting, and our numbers do not adjust for that. See what AI bias means and what these tests can tell you.
Best-supported use cases
This model is not among the options we recommend for any of the five tasks. That reflects the rules and the models listed, not a verdict on the model.
Where an alternative might be more suitable
- Writing and creating content: rated Middle on writing votes. Compare it with Claude Opus 5.5, the best-supported choice for this task.
- Coding and building things: rated Strong on coding votes. Compare it with Claude Opus 5.5, the best-supported choice for this task.
- Homework and maths: rated Middle on hard maths problems. Compare it with GPT-6.1 Sol, the best-supported choice for this task.
- Business emails and everyday work: rated Middle on business-task votes. Compare it with Gemini 4 Argon, the best-supported choice for this task.
- Research and checking facts: rated Behind the leaders on short factual questions. Compare it with Gemini 4 Argon, the best-supported choice for this task.
What it is
- Maker
- Moonshot
- Release date (Epoch AI)
- 20 Apr 2026
- Access, as listed by Epoch AI
- Open weights (unrestricted)
- Licence, as listed by Arena
- Modified MIT
- Maker's country (Epoch AI)
- China
- Last observation in our data
- 28 Oct 2026
- Data last changed
- 29 Oct 2026, 14:36 UTC
We have not verified an official product page for this model, so we do not link to one.
How you can use it
Open weights. Arena lists its licence as Modified MIT. In plain terms, the model's files are published under a licence that lets others run it, so it can be offered by more than one company, sometimes at lower cost. That does not guarantee that any particular app is free. Open licences differ, and some restrict commercial use, so check the licence terms before building a business on it.
How you can reach it (an app, a website, a developer API or a third-party tool) and what that costs are not covered by our benchmark data. Where we could verify prices from the maker's own pages, they are shown under Price. Epoch AI lists access as “Open weights (unrestricted)”. See models, apps and harnesses and how AI pricing works.
Price
Price not independently verified
We have not been able to check this maker's own pricing page, so we show nothing for it.
We never infer subscription prices from API prices. New to “tokens” and “per million”? See how AI pricing works.
Good to remember
- Benchmarks are samples of tasks. None proves performance in every situation.
- Arena scores come from people choosing between answers. That measures preference, not correctness.
- Ratings are relative to the 77 models listed here. A different set would change them.
- Sources test models at different settings. Each result shows the setting used, and we do not adjust it.
The evidence, for the curious
Show all 13 tests with settings and sources
| Test | Result | Rating | Setting used | Source identifier | Observed | Links |
|---|---|---|---|---|---|---|
| Arena text: overall preference Arena 'text' leaderboard (Bradley-Terry rating, no style control) |
1,456 Arena score 95% range 1,451 to 1,460, 39,978 votes |
Strong | effort: not stated, bradley-terry rating | Kimi K2.6 lmarena · text/latest |
28 Oct 2026 | Source for Kimi K2.6 Arena text: overall preference score (lmarena) |
| Arena text: creative writing Arena 'text' leaderboard, category creative_writing |
1,433 Arena score 95% range 1,425 to 1,441, 6,297 votes |
Middle | effort: not stated, bradley-terry rating | Kimi K2.6 lmarena · text/latest |
28 Oct 2026 | Source for Kimi K2.6 Arena text: creative writing score (lmarena) |
| Arena text: coding Arena 'text' leaderboard, category coding |
1,489 Arena score 95% range 1,482 to 1,496, 11,084 votes |
Strong | effort: not stated, bradley-terry rating | Kimi K2.6 lmarena · text/latest |
28 Oct 2026 | Source for Kimi K2.6 Arena text: coding score (lmarena) |
| Arena text: maths Arena 'text' leaderboard, category math |
1,474 Arena score 95% range 1,461 to 1,487, 2,075 votes |
Strong | effort: not stated, bradley-terry rating | Kimi K2.6 lmarena · text/latest |
28 Oct 2026 | Source for Kimi K2.6 Arena text: maths score (lmarena) |
| Arena text: following instructions Arena 'text' leaderboard, category instruction_following |
1,451 Arena score 95% range 1,445 to 1,457, 13,184 votes |
Middle | effort: not stated, bradley-terry rating | Kimi K2.6 lmarena · text/latest |
28 Oct 2026 | Source for Kimi K2.6 Arena text: following instructions score (lmarena) |
| Arena text: business and management Arena 'text' leaderboard, category industry_business_and_management_and_financial_operations |
1,445 Arena score 95% range 1,438 to 1,452, 7,923 votes |
Middle | effort: not stated, bradley-terry rating | Kimi K2.6 lmarena · text/latest |
28 Oct 2026 | Source for Kimi K2.6 Arena text: business and management score (lmarena) |
| Arena WebDev: building web apps Arena 'webdev' leaderboard (Code Arena), category overall |
1,509 Arena score 95% range 1,502 to 1,516, 10,119 votes |
Middle | effort: not stated, bradley-terry rating | Kimi K2.6 lmarena · webdev/latest |
28 Oct 2026 | Source for Kimi K2.6 Arena WebDev: building web apps score (lmarena) |
| FrontierMath Tiers 1-3 (v2) FrontierMath Tiers 1-3 v2 (private set), benchmark released 2026-06-12 |
57.2% share of problems solved standard error ±2.9 points |
Middle | effort: not stated, best score across scorers | kimi-k2.6 epoch · frontiermath_tiers_1_3_v2.csv |
10 Jun 2026 | Source for Kimi K2.6 FrontierMath Tiers 1-3 (v2) score (epoch) · Run details for FrontierMath Tiers 1-3 (v2) |
| OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025, benchmark released 2024-12-19 |
96.1% share of problems solved standard error ±2.4 points |
Middle | effort: not stated, best score across scorers | kimi-k2.6 epoch · otis_mock_aime_2024_2025.csv |
2 May 2026 | Source for Kimi K2.6 OTIS Mock AIME 2024-2025 score (epoch) · Run details for OTIS Mock AIME 2024-2025 |
| GPQA Diamond GPQA Diamond, benchmark released 2023-11-20 |
90.8% share of questions correct standard error ±1.7 points |
Strong | effort: not stated, best score across scorers | kimi-k2.6 epoch · gpqa_diamond.csv |
1 May 2026 | Source for Kimi K2.6 GPQA Diamond score (epoch) · Run details for GPQA Diamond |
| SimpleQA Verified SimpleQA Verified, benchmark released 2025-09-09 |
34.9% share of questions correct standard error ±1.5 points |
Behind the leaders | effort: not stated, best score across scorers | kimi-k2.6 epoch · simpleqa_verified.csv |
10 Aug 2026 | Source for Kimi K2.6 SimpleQA Verified score (epoch) No public run log in the data file. |
| SWE-bench Verified (Epoch runs) SWE-bench Verified, benchmark released 2024-08-13, as run by Epoch AI |
76.7% share of issues resolved standard error ±1.9 points |
Strong | effort: not stated, best score across scorers | kimi-k2.6 epoch · swe_bench_verified.csv |
8 May 2026 | Source for Kimi K2.6 SWE-bench Verified (Epoch runs) score (epoch) · Run details for SWE-bench Verified (Epoch runs) |
| FrontierCode (reported by Cognition) | Not enough data | |||||
Missing evidence
Not enough data for these tests:
- FrontierCode (reported by Cognition)
Missing results are not low scores and do not count against the model.
Where a source gives no public run log, we cannot show the run behind the score. See data sources.
Not sure this is the right one?
Tell us what you need help with and we'll narrow the options down.