Google model product
1 Pro is the latest model in the Gemini 3 series.
Updated Aug 12, 2026. Default version: Gemini 3.1 Pro
Technical details for the model's default version.
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Benchmark scores for Gemini 3.1 Pro.
| LiveCodeBench Pro | 2,887 points | 01 | 4 | 100.0% | C | |
| t2-bench | 99.3% | 01 | 23 | 100.0% | C | |
| ARC-AGI v2 | 77.1% | 02 | 16 | 93.3% | C | |
| MMMLU | 92.6% | 02 | 49 | 97.9% | C | |
| SciCode | 59.0% | 02 | 19 | 94.4% | C | |
| APEX-Agents | 33.5% | 03 | 7 | 66.7% | C | |
| GPQA | 94.3% | 03 | 234 | 99.1% | C | |
| LiveBench | 79.9% | 04 | 38 | 91.9% | B | |
| BrowseComp | 85.9% | 09 | 58 | 86.0% | C | |
| SWE-Bench Verified | 80.6% | 09 | 105 | 92.3% | C | |
| FrontierSWE | 40.0% | 10 | 15 | 35.7% | B | |
| Legal Agent Benchmark | 0.0% | 12 | 13 | 8.3% | B | |
| MMMU-Pro | 80.5% | 12 | 66 | 83.1% | C | |
| Terminal-Bench 2.0 | 68.5% | 12 | 49 | 77.1% | C | |
| Finance Agent v2 | 43.0% | 13 | 26 | 52.0% | B | |
| MRCR v2 (8-needle) | 26.3% | 15 | 21 | 30.0% | C | |
| Humanity's Last Exam | 51.4% | 17 | 93 | 82.6% | C | |
| MCP Atlas | 69.2% | 19 | 31 | 40.0% | C | |
| DeepSWE 1.1 | 12.0% | 20 | 20 | 0.0% | B | |
| SWE-Bench Pro | 54.2% | 34 | 45 | 25.0% | C |
Preference and agent-evaluation results for the default version.
| search style control | overall | 05 | 1209.7 | 89,973 | N/A | |
| search | overall | 07 | 1212.4 | 89,973 | N/A | |
| search factuality | overall | 11 | 1199.0 | 89,924 | N/A | |
| text | overall | 11 | 1479.3 | 94,409 | N/A | |
| text style control | overall | 13 | 1486.0 | 94,409 | N/A | |
| vision | overall | 18 | 1294.1 | 36,893 | N/A | |
| agent steerability | overall | 19 | 0.0 | N/A | 88.6K | |
| agent praise complaint | overall | 20 | 0.0 | N/A | 22.5K | |
| text factuality | overall | 20 | 1471.5 | 94,046 | N/A | |
| agent task outcome explicit | overall | 22 | 0.0 | N/A | 62.1K | |
| document style control | overall | 22 | 1459.8 | 45,162 | N/A | |
| document | overall | 23 | 1445.3 | 45,162 | N/A | |
| vision style control | overall | 25 | 1276.9 | 36,893 | N/A | |
| agent tool hallucination | overall | 26 | 0.0 | N/A | 2.1M | |
| agent | overall | 28 | -0.0 | N/A | 2.5M | |
| agent bash recovery steps | overall | 38 | -0.1 | N/A | 177.5K | |
| webdev | overall | 43 | 1446.7 | 20,473 | N/A |
Provider-specific output speed and catalog latency for Gemini 3.1 Pro. Runtime does not affect the capability score.
| 28.427 tok/s | 128.56 s | 1M | 65.5K |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Official vendor API pricing appears first, followed by individual provider offers.
| Impossibl | google/gemini-3.1-pro-preview | global | $2 | $12 | 1M | |
| Poe | google/gemini-3.1-pro | global | $2 | $12 | 1M | |
| Vivgrid | gemini-3.1-pro-preview | global | $2 | $12 | 1M | |
| gemini-3.1-pro-preview | global | $2 | $12 | 1M | ||
| NanoGPT | google/gemini-3.1-pro-preview | global | $2 | $12 | 1M | |
| FastRouter | google/gemini-3.1-pro-preview | global | $2 | $12 | 1M | |
| DaoXE | gemini-3.1-pro-preview | global | $2 | $12 | 1M | |
| LLM Gateway | gemini-3.1-pro-preview | global | $2 | $12 | 1M | |
| GitHub Copilot | gemini-3.1-pro-preview | global | $2 | $12 | 1M | |
| ZenMux | google/gemini-3.1-pro-preview | global | $2 | $12 | 1M | |
| Abacus | gemini-3.1-pro-preview | global | $2 | $12 | 1M | |
| Auriko | gemini-3.1-pro-preview | global | $2 | $12 | 1M | |
| AIHubMix | gemini-3.1-pro-preview | global | $2 | $12 | 1M | |
| Ofox | google/gemini-3.1-pro-preview | global | $2 | $12 | 1M | |
| Vertex | gemini-3.1-pro-preview | global | $2 | $12 | 1M | |
| Kilo Gateway | google/gemini-3.1-pro-preview | global | $2 | $12 | 1M | |
| Merge Gateway | google/gemini-3.1-pro-preview | global | $2 | $12 | 1M | |
| Vercel AI Gateway | google/gemini-3.1-pro-preview | global | $2 | $12 | 1M | |
| OpenRouter | google/gemini-3.1-pro-preview | global | $2 | $12 | 1M | |
| Perplexity Agent | google/gemini-3.1-pro-preview | global | $2 | $12 | 1M | |
| OrcaRouter | google/gemini-3.1-pro-preview | global | $4 | $18 | 1M |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Available versions of this model. The score column identifies the version used in the overall ranking.
| Gemini 3.1 Pro | 77.5 | N/A | 1M | 65.5K | No | Proprietary |
Key information about Gemini 3.1 Pro and its available data.
1 Pro is the latest model in the Gemini 3 series. It excels at complex tasks requiring broad world knowledge and advanced reasoning across modalities. 1 Pro uses dynamic thinking by default to reason through prompts, and features a 1 million-token input context window with 64k output tokens.
Data as of 2026-08-11.
Open a comparison with the three ranked models immediately above and below this model.
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
Common questions about Gemini 3.1 Pro.
Gemini 3.1 Pro's default version was released on Feb 19, 2026.
Gemini 3.1 Pro's official API price is $2 per million input tokens and $12 per million output tokens via Google. The lowest tracked third-party offer starts at $2 input and $12 output via Impossibl.
Gemini 3.1 Pro was created by Google.
The default version has a 1M token context window.
No. The default version is not marked as having publicly available weights.
21 provider offerings are linked to the default version.
Nearby ranked alternatives include Qwen3.7 Max, Seed 2.1 Pro, DeepSeek-V4-Flash.