Gemini 3.1 Pro vs Gemini 3.7 Flash
Gemini 3.1 Pro (Google DeepMind) and Gemini 3.7 Flash (Google DeepMind) compared on benchmarks, pricing, context window, and use-case rankings.
- Gemini 3.7 Flash leads on overall intelligence (ECI 157 vs 155).
- Gemini 3.7 Flash ranks higher for coding (94th vs 59th percentile).
| Attribute | Gemini 3.1 Pro | Gemini 3.7 Flash |
|---|---|---|
| Scores | ||
| Intelligence (ECI) | 155 | 157 |
| Coding | 59 | 94 |
| Math | 71 | 82 |
| Reasoning & Knowledge | 93 | 97 |
| Agentic & Tools | 79 | — |
| Specifications | ||
| Developer | Google DeepMind | Google DeepMind |
| Family | Gemini | Gemini |
| Released | Feb 19, 2026 | Aug 13, 2026 |
| Parameters | — | — |
| Availability | API access | API access |
| Context window | — | 1M |
| Price — $/M input | — | $0.75 |
| Price — $/M output | — | $3.75 |
| Inputs | — | text, image, video, file, audio |
| Outputs | — | text |
| Benchmarks | ||
| AIME 2024/2025 | 96% | 97% |
| APEX | 34% | — |
| ARC-AGI | 98% | — |
| ARC-AGI-2 | 77% | — |
| FrontierMath | 37% | — |
| FrontierMath Tier 4 | 17% | — |
| GPQA Diamond | 94% | 95% |
| GSO (code optimization) | 23% | — |
| Humanity's Last Exam | 46% | 48% |
| METR task horizon | 6.4 h | — |
| SimpleBench | 80% | — |
| SimpleQA Verified | 77% | 71% |
| SWE-bench Verified | 76% | — |
| Terminal-Bench | 80% | — |
| WebDev Arena | 1461 | 1587 |
| WeirdML | 72% | — |
| SciCode | — | 57% |
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 Gemini 3.1 Pro better than Gemini 3.7 Flash?
On Epoch AI's Capabilities Index, Gemini 3.7 Flash scores higher (157) than Gemini 3.1 Pro (155). The right pick depends on your task — compare their coding, math, and reasoning scores in the table above.
Which is better for coding, Gemini 3.1 Pro or Gemini 3.7 Flash?
Across coding benchmarks like SWE-bench Verified and Terminal-Bench, Gemini 3.7 Flash ranks higher — 94th vs 59th 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.