Model
Kimi K2.5
Moonshot AI · open weights
Knowledge & Reasoning
Knowledge & Reasoning| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| CL-bench | 19.3 | overall | N/A | Independent | low | epoch.ai |
| CL-bench Life | 13.2 | overall | N/A | Independent | low | epoch.ai |
| Colombian Legal Reliability Benchmark | 75.4 | accuracy | N/A | Independent | medium | arxiv.org |
| CritPt | 3.1 | accuracy | N/A | Independent | low | epoch.ai |
| EnigmaEval | 3.4 | accuracy | N/A | Independent | medium | labs.scale.com |
| GPQA Diamond | 87.6 | accuracy | N/A | Self-reported | low | artificialanalysis.ai |
| HLE | 24.4 | accuracy | N/A | Independent | low | epoch.ai |
| LMArena | 1438.0 | elo | N/A | Self-reported | low | lmarena.ai |
| MMLU-Pro | 87.1 | accuracy | N/A | Self-reported | low | artificialanalysis.ai |
| SimpleQA Verified | 34.3 | accuracy | N/A | Independent | high | epoch.ai |
Math
Math| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| AIME 2025 | 96.1 | accuracy | N/A | Self-reported | low | kaggle.com |
| MATH-500 | 98.0 | accuracy | N/A | Self-reported | low | kaggle.com |
Coding
Coding| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| LiveCodeBench | 85.0 | pass@1 | N/A | Self-reported | low | livecodebench.github.io |
| SciCode | 49.0 | score | N/A | Independent | medium | artificialanalysis.ai |
| SWE-bench Verified | 76.8 | resolved_rate | N/A | Self-reported | low | swebench.com |
| SWE-bench Verified | 73.8 | resolved_rate | N/A | Independent | high | epoch.ai |
| SWE-CC | 74.6 | resolve_rate | mini-swe-agent | Independent | medium | arxiv.org |
| Terminal-Bench 2.0 | 50.8 | accuracy | N/A | Self-reported | low | tbench.ai |
Agentic / Tool Use
Agentic / Tool Use| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| 2D-FET-Bench V2 | 23.0 | pass@1 | react-3/c0 | Independent | medium | arxiv.org |
| EHR-RobustGym | 40.5 | success_rate | N/A | Independent | medium | arxiv.org |
| OSWorld | 63.3 | score | Kimi-K2.5 | Independent | medium | os-world.github.io |
| Terminal-Bench | 43.2 | accuracy | Terminus 2 | Independent | medium | tbench.ai |
| XiangqiBench | 1.7 | pass@3 | N/A | Independent | medium | arxiv.org |
Multimodal
Multimodal| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| ChartBmkAgent | 63.0 | accuracy | N/A | Independent | medium | arxiv.org |
| DSV-Mem | 5.8 | accuracy | N/A | Independent | medium | arxiv.org |
| EditJudgeBias | 5.2 | rubric_ratings_(1-10)_and_pairwise_preferences | N/A | Independent | medium | arxiv.org |
| PhysVista | 90.0 | order_accuracy | N/A | Independent | medium | arxiv.org |
Instruction Following
Instruction Following| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| IFEval | 94.0 | prompt_strict_accuracy | N/A | Self-reported | low | artificialanalysis.ai |
Safety & Alignment
Safety & Alignment| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| Live-Harness-Bench | 8.4 | success_rate | live-harness-bench | Independent | medium | arxiv.org |
Abstract Reasoning
Abstract ReasoningReproduced by the community
Loading community runs…