Gemini 3.1 Pro Preview
74.2#30 of 427 overall
- Input / output
- US$ 2,00 / US$ 12,00 US$ per 1M tokens
- Context
- 1.0M tokens
- Release
- 02/19/2026 release date
- Weights
- Closed
- Speed
- 141 tok/s first token in 15.04 s
Family history
The 7 versions of this family, from first to newest, by release date. Click a point to open that version.
See the data as a table
| Model | Released | Score | Rank | Elo Arena |
|---|---|---|---|---|
| Gemini 1.0 Pro | 12/06/2023 | 32.5 | #327 | — |
| Gemini 1.5 Pro | 09/24/2024 | 26.4 | #381 | 1319 |
| Gemini 2.0 Pro Experimental | 02/05/2025 | 44.3 | #202 | — |
| Gemini 2.5 Pro | 06/17/2025 | 52.8 | #124 | 1458 |
| Gemini 3 Pro Preview | 11/18/2025 | 74.4 | #28 | 1479 |
| Gemini 3.1 Pro Preview | 02/19/2026 | 74.2 | #30 | 1480 |
| Gemini Pro Latest | 04/27/2026 | — | — | 1149 |
Scores by source
Scores by sourcepercentile = % of ranked models it beats| Metric | Value | Percentile | Source |
|---|---|---|---|
| Arena text (Elo) weight 2 | 1480 | 92% | LMArena |
| Arena WebDev (Elo) weight 1 | 1447 | 47% | LMArena |
| Arena vision (Elo) weight 0.5 | 1296 | 84% | LMArena |
| Humanity's Last Exam weight 1 | 47.0 | 96% | Artificial Analysis |
| Terminal-Bench weight 1 | 4.0 | 65% | Artificial Analysis |
| GPQA Diamond | 94.1 | 99% | Artificial Analysis |
| Long-context reasoning | 82.0 | 92% | Artificial Analysis |
By category
| Category | Score | Rank | Metrics |
|---|---|---|---|
| Agentic | 57.5 | #43 | Terminal-Bench |
| Coding | 48.3 | #61 | Arena WebDev (Elo) |
| Reasoning | 81.5 | #8 | Humanity's Last Exam, GPQA Diamond |
| Research | 71.0 | #25 | Long-context reasoning |
| Design | 48.3 | #61 | Arena WebDev (Elo) |
| Chat | 70.8 | #18 | Arena text (Elo) |
| Vision | 66.9 | #18 | Arena vision (Elo) |
History
RankingIA score on each collection day (hover to see the rank).
Daily collection started on 10/30/2026. The chart appears from the second day the model is ranked.