Compare
Gemini 3 Pro vs Kimi K3
Add, remove, or swap models to compare them side by side.
| Attribute | Gemini 3 Pro | Kimi K3 |
|---|---|---|
| Scores | ||
| Intelligence (ECI) | 153 | 157 |
| Coding | 51 | 99 |
| Math | 68 | 83 |
| Reasoning & Knowledge | 74 | 79 |
| Agentic & Tools | 57 | 86 |
| Specifications | ||
| Developer | Google DeepMind | Moonshot |
| Family | Gemini | Kimi |
| Released | Nov 18, 2025 | Jul 16, 2026 |
| Parameters | — | 2.8T |
| Availability | API access | Open weights (non-commercial) |
| Context window | — | 1M |
| Price — $/M input | — | $3.00 |
| Price — $/M output | — | $14.00 |
| Inputs | — | text, image |
| Outputs | — | text |
| Benchmarks | ||
| AIME 2024/2025 | 91% | 97% |
| APEX | 32% | 39% |
| ARC-AGI | 75% | 95% |
| ARC-AGI-2 | 31% | 60% |
| FrontierMath | 38% | — |
| FrontierMath Tier 4 | 19% | — |
| GDPval (win/tie rate) | 54% | — |
| GPQA Diamond | 93% | 93% |
| GSO (code optimization) | 19% | — |
| Humanity's Last Exam | 38% | 44% |
| METR task horizon | 3.7 h | — |
| SimpleBench | 76% | — |
| SimpleQA Verified | 73% | 43% |
| SWE-bench Verified | 73% | — |
| Terminal-Bench | 69% | — |
| WebDev Arena | 1471 | 1682 |
| WeirdML | 70% | 83% |
| SciCode | — | 59% |
Use-case scores are 0–100 percentile composites across each area’s benchmarks, ranked against every model from the past year. Highlighted cells lead each row. Open a model for the full picture.
Popular comparisons
- Claude Fable 5 vs GPT-5.6 Sol
- GPT-5.6 Sol vs Kimi K3
- Gemini 3.1 Pro vs GPT-5.6 Sol
- GPT-5.6 Sol vs Grok 4.5
- GPT-5.6 Sol vs Qwen 3.7 Max
- GPT-5.6 Sol vs Muse Spark
- GLM-5.2 vs GPT-5.6 Sol
- Claude Fable 5 vs Kimi K3
- Claude Fable 5 vs Gemini 3.1 Pro
- Claude Fable 5 vs Grok 4.5
- Claude Fable 5 vs Qwen 3.7 Max
- Claude Fable 5 vs Muse Spark
- Claude Fable 5 vs GLM-5.2
- Gemini 3.1 Pro vs Kimi K3
- Grok 4.5 vs Kimi K3
- Kimi K3 vs Qwen 3.7 Max
- Kimi K3 vs Muse Spark
- GLM-5.2 vs Kimi K3
- Gemini 3.1 Pro vs Grok 4.5
- Gemini 3.1 Pro vs Qwen 3.7 Max
- Gemini 3.1 Pro vs Muse Spark
- Gemini 3.1 Pro vs GLM-5.2
- Grok 4.5 vs Qwen 3.7 Max
- Grok 4.5 vs Muse Spark
- GLM-5.2 vs Grok 4.5
- Muse Spark vs Qwen 3.7 Max
- GLM-5.2 vs Qwen 3.7 Max
- GLM-5.2 vs Muse Spark
- Claude Fable 5 vs Claude Opus 5
- Kimi K2 vs Kimi K3
- GLM-5 vs GLM-5.2
- GPT-5.5 vs GPT-5.6 Sol
- Kimi K2.5 vs Kimi K3
- GPT-5 vs GPT-5.6 Sol
Benchmarks & model data from Epoch AI (CC BY); pricing & specs from OpenRouter. ECI = Epoch Capabilities Index.