Kimi K2.6
by Moonshot AI · ChinaMoonshot's April 2026 open-weight release, with thinking and non-thinking modes and video input. Moonshot's own table put it ahead of Claude Opus 4.6 and GPT-5.4 on SWE-bench Pro.
Specification
The numbers
- Maker
- Moonshot AI
- Released
- 2026-04
- Parameters
- Not reported
- Architecture
- MoE
- Context window
- 262,144 tokens
- Max output
- Not reported
- Input
- Text, Image, Video
- Output
- Text
- Reasoning
- Yes
- Tool calling
- Yes
- Knowledge cutoff
- Not reported
- Licence
- Not stated
- Availability
- Not reported
- Weights
- Downloadable
Cost
Price per million tokens
No public API price. The weights are published, so the only cost is the hardware you run it on.
Published scores
Benchmarks
Figures published by Moonshot AI or taken from a public leaderboard. Row last checked 2026-10.
| Category | Score | Rank |
|---|---|---|
| ReasoningGPQA Diamond | 88.4% | 17 of 99 reporting the same tests |
| MathsCompetition mathematics, graded on the final answer | Not reported | — |
| CodingSWE-bench Pro | 58.6% | 10 of 14 reporting the same tests |
| KnowledgeBreadth of factual recall under exam conditions | Not reported | — |
| MultimodalReading charts, diagrams and photographs | Not reported | — |
| Instruction followingObeying an exact, checkable format | Not reported | — |
| Human preferenceWhich answer people pick, blind | Not reported | — |
A category averages every benchmark in it that Kimi K2.6 reports. The rank counts only models that report the same tests, so it never compares an average over three benchmarks against an average over one.
Every reported test
| MMLU-ProKnowledge | Not reported | No figure published |
|---|---|---|
| GPQA DiamondReasoning | Rank 17 of 99 models reporting | |
| AIME 2025Maths | Not reported | No figure published |
| MATH-500Maths | Not reported | No figure published |
| SWE-bench VerifiedCoding | Not reported | No figure published |
| SWE-bench ProCoding | Rank 10 of 14 models reporting | |
| Terminal-Bench 2.0Coding | Not reported | No figure published |
| Terminal-Bench 2.1Coding | Not reported | No figure published |
| Frontier-Bench v0.1Reasoning | Not reported | No figure published |
| Terminal-Bench 4.0Coding | Not reported | No figure published |
| LiveCodeBenchCoding | Not reported | No figure published |
| HumanEvalCoding | Not reported | No figure published |
| MMMUMultimodal | Not reported | No figure published |
| IFEvalInstruction following | Not reported | No figure published |
| LMArena EloHuman preference | Not reported | No figure published |
Where these numbers come from
Every score on this page is a published figure, taken from the model's own card, system card, technical report or release post, or from a public leaderboard. CorX Labs did not run these evaluations. Most are self-reported by the lab that built the model, which means they were produced under that lab's own choice of prompt, scaffold and number of attempts — so treat them as a starting point for a shortlist, not as a settled ranking.
A score someone other than the model's maker measured is marked Independent and names its measurer. Those are the stronger numbers on this page — an outside harness has no reason to flatter anyone — and there are not many of them.
Where a figure has not been published, the cell reads Not reported rather than an estimate. Nothing here is inferred, interpolated or guessed. Each model records the month its row was last checked. Full method and caveats.
Head to head
Kimi K2.6 compared
Same maker
Other models from Moonshot AI
| Kimi K3Moonshot AI | Moonshot AI | 1M | $3.00 | $15.00 | — | — | — | — | Kimi K3 License |
| Kimi K2 ThinkingMoonshot AI | Moonshot AI | 262K | $0.60 | $2.50 | — | 84.5% | 94.5% | 71.3% | Modified MIT |
| Kimi K2 InstructMoonshot AI | Moonshot AI | 131K | $0.60 | $2.50 | 81.1% | — | — | 65.8% | Modified MIT |
| Kimi-Dev-72BMoonshot AI | Moonshot AI | 131K | $0.290 | $1.15 | — | — | — | 60.4% | Modified MIT |