Claude Sonnet 4.5 vs Gemini 3.1 Pro
Claude Sonnet 4.5 (Anthropic) and Gemini 3.1 Pro (Google DeepMind) compared on benchmarks, pricing, context window, and use-case rankings.
- Gemini 3.1 Pro leads on overall intelligence (ECI 155 vs 147).
- Gemini 3.1 Pro ranks higher for coding (65th vs 34th percentile).
| Attribute | Claude Sonnet 4.5 | Gemini 3.1 Pro |
|---|---|---|
| Scores | ||
| Intelligence (ECI) | 147 | 155 |
| Coding | 34 | 65 |
| Math | 32 | 74 |
| Reasoning & Knowledge | 31 | 94 |
| Agentic & Tools | 39 | 83 |
| Specifications | ||
| Developer | Anthropic | Google DeepMind |
| Family | Claude | Gemini |
| Released | Sep 29, 2025 | Feb 19, 2026 |
| Parameters | — | — |
| Availability | API access | API access |
| Context window | 1M | — |
| Price — $/M input | $3.00 | — |
| Price — $/M output | $15.00 | — |
| Inputs | text, image, file | — |
| Outputs | text | — |
| Benchmarks | ||
| AIME 2024/2025 | 78% | 96% |
| ARC-AGI | 64% | 98% |
| ARC-AGI-2 | 14% | 77% |
| FrontierMath | 15% | 37% |
| FrontierMath Tier 4 | 4% | 17% |
| GDPval (win/tie rate) | 50% | — |
| GPQA Diamond | 82% | 94% |
| GSO (code optimization) | 15% | 23% |
| Humanity's Last Exam | 14% | 46% |
| MATH Level 5 | 98% | — |
| METR task horizon | 2.0 h | 6.4 h |
| SimpleBench | 54% | 80% |
| SimpleQA Verified | 24% | 77% |
| SWE-bench Verified | 71% | 76% |
| Terminal-Bench | 47% | 80% |
| WebDev Arena | 1391 | 1461 |
| WeirdML | 48% | 72% |
| APEX | — | 34% |
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.
Frequently asked questions
Is Claude Sonnet 4.5 better than Gemini 3.1 Pro?
On Epoch AI's Capabilities Index, Gemini 3.1 Pro scores higher (155) than Claude Sonnet 4.5 (147). The right pick depends on your task — compare their coding, math, and reasoning scores in the table above.
Which is better for coding, Claude Sonnet 4.5 or Gemini 3.1 Pro?
Across coding benchmarks like SWE-bench Verified and Terminal-Bench, Gemini 3.1 Pro ranks higher — 65th vs 34th percentile among the models tracked on Model Beat.
Want a different match-up? Open the compare tool to add or swap models.
More comparisons
Benchmarks & model data from Epoch AI (CC BY); pricing & specs from OpenRouter. ECI = Epoch Capabilities Index.