# Kimi K2.6 Benchmark Scores & Performance

> Kimi K2.6 by Moonshot AI scores 58.74/100 overall, ranking #53 out of 889 AI models.

Canonical page: https://benchlm.ai/models/kimi-2-6

Last updated: October 10, 2026

## Model Details

| Property | Value |
|----------|-------|
| Creator | Moonshot AI |
| Source Type | Open Weight |
| Reasoning Type | Reasoning |
| Context Window | 256K |
| Overall Score | 58.74/100 |
| Overall Rank | #53 of 889 |

## Family & Coverage

- Family: Kimi K2.6
- Variant: base
- Benchmarks covered: 55 of 625
- Related earlier model: [Kimi K2.5](/models/kimi-k2-5)
- Coverage note: BenchLM currently has partial benchmark coverage for this model, so the overall score is conservative.

## Agentic Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [Terminal-Bench 2.0](/benchmarks/terminal-bench-2) | 66.7% | [MoonshotAI: Kimi K2.6 tech blog](https://www.kimi.com/blog/kimi-k2-6) |
| [BrowseComp](/benchmarks/browsecomp) | 83.2% | [MoonshotAI: Kimi K2.6 tech blog](https://www.kimi.com/blog/kimi-k2-6) |
| [OSWorld-Verified](/benchmarks/osworld-verified) | 73.1% | [MoonshotAI: Kimi K2.6 tech blog](https://www.kimi.com/blog/kimi-k2-6) |
| [Toolathlon](/benchmarks/toolathlon) | 50% | [MoonshotAI: Kimi K2.6 tech blog](https://www.kimi.com/blog/kimi-k2-6) |
| [MCP Atlas](/benchmarks/mcpatlas) | 55.9% | [MoonshotAI: Kimi K2.6 model card](https://huggingface.co/moonshotai/Kimi-K2.6) |
| [Claw-Eval](/benchmarks/claw-eval) | 62.3% | [Claw-Eval leaderboard](https://claw-eval.github.io/) |
| [DeepSearchQA](/benchmarks/deepsearchqa) | 92.5% | [MoonshotAI: Kimi K2.6 tech blog](https://www.kimi.com/blog/kimi-k2-6) |
| [WideResearch](/benchmarks/wideresearch) | 80.8% | [MoonshotAI: Kimi K2.6 model card](https://huggingface.co/moonshotai/Kimi-K2.6) |
| [τ²-bench results](/benchmarks/tau2-bench) | 95.9% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/kimi-k2-6) |
| [GDPval-AA](/benchmarks/gdpvalaanormalized) | 27.0% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/kimi-k2-6) |
| [APEX-Agents-AA](/benchmarks/apexagentsaa) | 28.5% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/kimi-k2-6) |
| [Gert Labs](/benchmarks/gertlabs) | 56.82% | [Gert Labs rankings](https://gertlabs.com/rankings) |
| [ResearchClawBench](/benchmarks/researchclawbench) | 18.0% | [ResearchClawBench leaderboard](https://internscience.github.io/ResearchClawBench-Home/) |
| [OSWorld 2.0](/benchmarks/osworld2) | 4.6% | [OSWorld 2.0 paper](https://arxiv.org/abs/2606.29537) |
| [Terminal-Bench 2.1 (Vals)](/benchmarks/terminalbench21) | 53.6% | [Vals AI: Terminal-Bench 2.1 leaderboard](https://www.vals.ai/models/kimi_kimi-k2.6) |
| [AA Agentic Index](/benchmarks/aaagenticindex) | 22.1% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/kimi-k2-6) |
| [GDPval-AA](/benchmarks/gdpvalaa) | 1115 | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/kimi-k2-6) |

## Coding Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [SWE-bench Verified](/benchmarks/swe-bench-verified) | 80.2% | [MoonshotAI: Kimi K2.6 model card](https://huggingface.co/moonshotai/Kimi-K2.6) |
| [LiveCodeBench v6](/benchmarks/livecodebench-v6) | 89.6% | [MoonshotAI: Kimi K2.6 model card](https://huggingface.co/moonshotai/Kimi-K2.6) |
| [SWE-bench Pro](/benchmarks/swe-bench-pro) | 58.6% | [MoonshotAI: Kimi K2.6 model card](https://huggingface.co/moonshotai/Kimi-K2.6) |
| [SWE Multilingual](/benchmarks/swe-bench-multilingual) | 76.7% | [MoonshotAI: Kimi K2.6 tech blog](https://www.kimi.com/blog/kimi-k2-6) |
| [SciCode](/benchmarks/scicode) | 52.2% | [MoonshotAI: Kimi K2.6 model card](https://huggingface.co/moonshotai/Kimi-K2.6) |
| [Terminal-Bench 2.0](/benchmarks/terminal-bench-2) | 66.7% | [MoonshotAI: Kimi K2.6 tech blog](https://www.kimi.com/blog/kimi-k2-6) |
| [Vibe Code Bench](/benchmarks/vibecodebench) | 37.89% | [Vals AI: Vibe Code Bench v1.1](https://www.vals.ai/benchmarks/vibe-code) |
| [cursorBench31](/benchmarks/cursorbench31) | 47.6% | [Cursor evals: CursorBench 3.1](https://cursor.com/evals) |
| [AA-SciCode](/benchmarks/aascicode) | 51.5% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/kimi-k2-6) |
| [LiveCodeBench (Vals)](/benchmarks/valslivecodebench) | 86.8% | [Vals AI: LiveCodeBench leaderboard](https://www.vals.ai/models/kimi_kimi-k2.6) |
| [SWE-bench (Vals)](/benchmarks/valsswebench) | 76.2% | [Vals AI: SWE-bench leaderboard](https://www.vals.ai/models/kimi_kimi-k2.6) |
| [AA Coding Index](/benchmarks/aacodingindex) | 61.8% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/kimi-k2-6) |

## Multimodal & Grounded Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [MMMU-Pro](/benchmarks/mmmu-pro) | 79.4% | [MoonshotAI: Kimi K2.6 model card](https://huggingface.co/moonshotai/Kimi-K2.6) |
| [MMMU-Pro w/ Python](/benchmarks/mmmupropython) | 80.1% | [MoonshotAI: Kimi K2.6 model card](https://huggingface.co/moonshotai/Kimi-K2.6) |
| [CharXiv](/benchmarks/charxiv) | 80.4% | [MoonshotAI: Kimi K2.6 model card](https://huggingface.co/moonshotai/Kimi-K2.6) |
| [MathVision](/benchmarks/mathvision) | 87.4% | [MoonshotAI: Kimi K2.6 model card](https://huggingface.co/moonshotai/Kimi-K2.6) |
| [V*](/benchmarks/vstar) | 96.9% | [MoonshotAI: Kimi K2.6 tech blog](https://www.kimi.com/blog/kimi-k2-6) |
| [AA-MMMU-Pro](/benchmarks/mmmu-pro) | 79.4% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/kimi-k2-6) |
| [Design Arena Website](/benchmarks/designarenawebsite) | 1274 | [OpenRouter model benchmarks](https://openrouter.ai/moonshotai/kimi-k2.6/benchmarks) |

## Reasoning Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [AA-LCR](/benchmarks/lcr) | 81.0% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/kimi-k2-6) |
| [CritPt](/benchmarks/critpt) | 8.0% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/kimi-k2-6) |

## Knowledge Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [GPQA](/benchmarks/gpqa) | 90.5% | [MoonshotAI: Kimi K2.6 model card](https://huggingface.co/moonshotai/Kimi-K2.6) |
| [GPQA-D](/benchmarks/gpqa-diamond) | 90.5% | [MoonshotAI: Kimi K2.6 model card](https://huggingface.co/moonshotai/Kimi-K2.6) |
| [HLE](/benchmarks/hle) | 34.7% | [MoonshotAI: Kimi K2.6 model card](https://huggingface.co/moonshotai/Kimi-K2.6) |
| [Artificial Analysis Intelligence Index](/benchmarks/artificialanalysis) | 27.0% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/kimi-k2-6) |
| [AA-GPQA Diamond](/benchmarks/gpqa-diamond) | 91.1% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/kimi-k2-6) |
| [AA-HLE](/benchmarks/aahle) | 37.5% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/kimi-k2-6) |
| [AA-Omniscience Index](/benchmarks/aaomniscienceindex) | 5.3% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/kimi-k2-6) |
| [AA-Omniscience Accuracy](/benchmarks/omniscienceaccuracy) | 32.6% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/kimi-k2-6) |
| [AA-Omniscience Hallucination Rate](/benchmarks/omnisciencehallucinationrate) | 40.5% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/kimi-k2-6) |
| [GPQA Diamond (Vals)](/benchmarks/gpqa-diamond) | 89.1% | [Vals AI: GPQA Diamond leaderboard](https://www.vals.ai/models/kimi_kimi-k2.6) |
| [MMLU-Pro (Vals)](/benchmarks/mmlu-pro) | 87.6% | [Vals AI: MMLU Pro leaderboard](https://www.vals.ai/models/kimi_kimi-k2.6) |

## Instruction Following Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [AA-IFBench](/benchmarks/ifbench) | 76.0% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/kimi-k2-6) |

## Mathematics Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [AIME26](/benchmarks/aime2026) | 96.4% | [MoonshotAI: Kimi K2.6 model card](https://huggingface.co/moonshotai/Kimi-K2.6) |
| [HMMT Feb 2026](/benchmarks/hmmtfeb2026) | 92.7% | [MoonshotAI: Kimi K2.6 model card](https://huggingface.co/moonshotai/Kimi-K2.6) |
| [MMAnswerBench](/benchmarks/mmanswerbench) | 86.0% | [MoonshotAI: Kimi K2.6 model card](https://huggingface.co/moonshotai/Kimi-K2.6) |
| [FrontierMath v2 (Tiers 1-3)](/benchmarks/frontiermathv2tiers13) | 38.966% | [Epoch AI FrontierMath v2 leaderboard](https://epoch.ai/benchmarks/frontiermath-tier-4-v2?view=graph&tab=leaderboard) |
| [FrontierMath v2 (Tier 4)](/benchmarks/frontiermathv2tier4) | 14.580% | [Epoch AI FrontierMath v2 leaderboard](https://epoch.ai/benchmarks/frontiermath-tier-4-v2?view=graph&tab=leaderboard) |

## Other Moonshot AI Models

- [Kimi K3](/models/kimi-k3) - Score: 70.8
- [Kimi K2.5 (Reasoning)](/models/kimi-k2-5-reasoning) - Score: 56.39
- [Kimi K2.7 Code](/models/kimi-k2-7-code) - Score: 52.39
- [Kimi K2.5](/models/kimi-k2-5) - Score: 52.27
- [Kimi K2](/models/kimi-k2) - Score: 34.03
- [Moonshot v1](/models/moonshot-v1) - Score: not computed
- [Kimi-Audio 7B](/models/kimi-audio-7b) - Score: not computed