Gemini 3.1 Pro Preview
- Context
- 1.05M
- 66K max output
- Input price
- $2.00
- per 1M tokens · Gemini API
- Output price
- $12.00
- per 1M tokens
- Epoch Capabilities Index
- 154.9
- #28 of 71 listed
- LMArena Text
- 1,487
- #13 of 79 listed
Where to run it
1 offering · prices as listed| Meter | Tier | Price |
|---|---|---|
| Input | Standard | $2.00 / 1M tokens |
| Batch | $1.00 / 1M tokens | |
| Flex tier | $1.00 / 1M tokens | |
| Priority tier | $3.60 / 1M tokens | |
| Prompts over 200K tokens | $4.00 / 1M tokens | |
| Prompts over 200K tokens · Batch | $2.00 / 1M tokens | |
| Prompts over 200K tokens · Flex tier | $2.00 / 1M tokens | |
| Prompts over 200K tokens · Priority tier | $7.20 / 1M tokens | |
| Output | Standard | $12.00 / 1M tokens |
| Batch | $6.00 / 1M tokens | |
| Flex tier | $6.00 / 1M tokens | |
| Prompts over 200K tokens | $18.00 / 1M tokens | |
| Priority tier | $21.60 / 1M tokens | |
| Prompts over 200K tokens · Batch | $9.00 / 1M tokens | |
| Prompts over 200K tokens · Flex tier | $9.00 / 1M tokens | |
| Prompts over 200K tokens · Priority tier | $32.40 / 1M tokens | |
| Cache read | Standard | $0.20 / 1M tokens |
| Batch | $0.20 / 1M tokens | |
| Flex tier | $0.20 / 1M tokens | |
| Priority tier | $0.36 / 1M tokens | |
| Prompts over 200K tokens | $0.40 / 1M tokens | |
| Prompts over 200K tokens · Batch | $0.40 / 1M tokens | |
| Prompts over 200K tokens · Flex tier | $0.40 / 1M tokens | |
| Prompts over 200K tokens · Priority tier | $0.72 / 1M tokens | |
Changes
All changes →- 4 new prices on Gemini API Input (Prompts over 200K tokens, Batch) $2.00 · Output (Prompts over 200K tokens, Batch) $9.00 · Input (Prompts over 200K tokens, Priority tier) $7.20 · Output (Prompts over 200K tokens, Priority tier) $32.40
Vendor-reported scores
Published by Google, under the conditions it states| Benchmark | Setting | Score |
|---|---|---|
| Humanity's Last Exam | full set, text + MM, no tools, thinking high | 44.4% |
| full set, text + MM, search (blocklist) + code, thinking high | 51.4% | |
| GPQA Diamond | no tools, pass@1, thinking high | 94.3% |
| Terminal-Bench 2.0 | Terminus-2 harness, thinking high | 68.5% |
| SWE-bench Verified | single attempt, averaged over 10 runs, thinking high | 80.6% |
| SWE-bench Pro | public set, single attempt, averaged over 5 runs, thinking high | 54.2% |
| τ²-bench | retail, thinking high | 90.8% |
| telecom, thinking high | 99.3% | |
| BrowseComp | search + Python + browse (Deep Research), thinking high | 85.9% |
| MMMU-Pro | no tools, average of Standard (10 options) and Vision, thinking high | 80.5% |
| MMMLU | thinking high | 92.6% |
Cited scores
From third-party leaderboards, not measured by us- Epoch Capabilities Index 154.9
#28 of 71 listed models · CI 152.8–157.6
Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Sep 26, 2026 · changes we made
- LMArena Text 1,487
#13 of 79 listed models · CI 1,484–1,490 · 119,196 votes
LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 25, 2026 · changes we made
- Epoch AI · FrontierMath Tiers 1–3 59.6%
#28 of 56 listed models · ±2.91
Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Sep 26, 2026 · changes we made
- Epoch AI · GPQA Diamond 94.4%
#5 of 64 listed models · high effort · ±1.63
Also listed: gemini-3.1-pro-preview 94.1%
Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Sep 26, 2026 · changes we made
- LMArena Text · Chinese 1,528
#16 of 79 listed models · CI 1,521–1,536 · 8,183 votes
LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 25, 2026 · changes we made
- LMArena Text · Coding 1,521
#21 of 79 listed models · CI 1,516–1,525 · 33,049 votes
LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 25, 2026 · changes we made
- LMArena Vision 1,296
#14 of 49 listed models · CI 1,290–1,301 · 40,691 votes
LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 13, 2026 · changes we made