Updated daily at 05:00 UTC
Vision-language
Gemini 3 Pro
- Released
- 18 Nov 2025
- Status
- Retired
- Retired
- 9 Mar 2026
- Context
- 1M
- Max output
- 65.5K
- Input per 1M tokens
- $2.00
- Output per 1M tokens
- $12.00
- Open weights
- No
- License
- proprietary
- Parameters
- —
- Knowledge cutoff
- Jan 2025
- Reasoning
- Yes
- Modalities
- text, image, video, audio, pdf → text
Category scores
Out of 100, from independent and third-party results only, by method category-2026-09-29.1.
| Category | Score | Rank | Coverage |
|---|---|---|---|
| Knowledge & reasoning | 49.5 | 17 | 2 of 4 |
| Coding & software engineering | 18.6 | Not ranked | 1 of 5 |
| Agentic tool use | 52.2 | 2 | 2 of 2 |
| Vision & multimodal understanding | 85.2 | 1 | 2 of 3 |
| Human preference | 52.4 | 24 | 4 of 5 |
Benchmark results
Every result we hold for this model, with who ran it and where it was published.
Knowledge & reasoning
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| Chess Puzzles | Puzzles solved | 31.0% | Independent | — | 8 Dec 2025 | Epoch AI |
| SimpleBench | Score (AVG@5) | 76.4% | Third party | — | — | https://lmcouncil.ai/benchmarks via Epoch AI |
Scientific reasoning
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| GPQA Diamond | Accuracy | 92.6% | Independent | — | 19 Nov 2025 | Epoch AI |
Mathematics
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| FrontierMath Tier 4 (v1) | Accuracy | 18.8% | Independent | — | 21 Nov 2025 | Epoch AI |
| FrontierMath Tiers 1-3 (v1) | Accuracy | 37.6% | Independent | — | 21 Nov 2025 | Epoch AI |
| OTIS Mock AIME 2024-2025 | Accuracy | 91.4% | Independent | — | 19 Nov 2025 | Epoch AI |
Coding & software engineering
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| GSO | OPT@1 | 18.6% | Third party | — | — | https://gso-bench.github.io/index.html via Epoch AI |
| METR Time Horizons | Average task success (Epoch's column) | 71.0% | Source not stated | — | — | Epoch AI via Epoch AI |
| SWE-bench Verified | % resolved | 72.9% | Independent | — | 13 Feb 2026 | Epoch AI |
| Terminal-Bench 2.0 | Accuracy (mean over trials) | 69.4% | Third party | — | — | https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI |
| Terminal-Bench 2.0 | Accuracy (mean over trials) | 65.2% | Third party | — | — | https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI |
| Terminal-Bench 2.0 | Accuracy (mean over trials) | 62.2% | Third party | — | — | https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI |
| Terminal-Bench 2.0 | Accuracy (mean over trials) | 61.8% | Third party | — | — | https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI |
| Terminal-Bench 2.0 | Accuracy (mean over trials) | 61.1% | Third party | — | — | https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI |
| Terminal-Bench 2.0 | Accuracy (mean over trials) | 56.9% | Third party | — | — | https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI |
| Terminal-Bench 2.0 | Accuracy (mean over trials) | 56.0% | Third party | — | — | https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI |
| Terminal-Bench 2.0 | Accuracy (mean over trials) | 62.2% | Source not stated | — | — | Epoch AI via Epoch AI |
| WeirdML (v2) | Average accuracy | 69.9% | Third party | — | — | https://htihle.github.io/weirdml.html via Epoch AI |
Agentic tool use
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| BALROG | Average progress | 58.1% | Third party | — | — | https://balrogai.com/ via Epoch AI |
| Deep Research Bench (FutureSearch) | Average score | 46.3% | Third party | low | — | https://evals.futuresearch.ai/#drb via Epoch AI |
Vision & multimodal understanding
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| GeoBench | Country accuracy | 84.0% | Third party | — | — | GeoBench leaderboard via Epoch AI |
| Visual Physics Comprehension Test (VPCT) | Correct | 91.0% | Third party | — | — | VPCT leaderboard via Epoch AI |
Human preference
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| Code Arena: WebDev (arena.ai) | Arena score (Elo scale) | 1439 | Independent | — | 25 Sep 2026 | Arena |
| Code Arena: WebDev (arena.ai) | Votes | 14103 | Independent | — | 25 Sep 2026 | Arena |
| Document Arena (arena.ai) | Arena score (Elo scale) | 1451 | Independent | — | 13 Sep 2026 | Arena |
| Document Arena (arena.ai) | Votes | 10769 | Independent | — | 13 Sep 2026 | Arena |
| Text Arena (arena.ai) | Arena score (Elo scale) | 1485 | Independent | — | 25 Sep 2026 | Arena |
| Text Arena (arena.ai) | Votes | 41921 | Independent | — | 25 Sep 2026 | Arena |
| Vision Arena (arena.ai) | Arena score (Elo scale) | 1289 | Independent | — | 28 Sep 2026 | Arena |
| Vision Arena (arena.ai) | Votes | 13624 | Independent | — | 28 Sep 2026 | Arena |
Other Gemini Pro versions
- Gemini 3.1 Pro19 Feb 2026
- Gemini 2.5 Pro25 Mar 2025
- Gemini 2.0 Pro6 Dec 2024
- Gemini 1.5 Pro15 Feb 2024
- Gemini 1.0 Pro Vision13 Dec 2023
- Gemini 1.0 Pro6 Dec 2023
Sources
- https://ai.google.dev/gemini-api/docs/models/gemini-3-pro-preview
- https://ai.google.dev/gemini-api/docs/deprecations
- https://ai.google.dev/gemini-api/docs/changelog
- https://ai.google.dev/gemini-api/docs/gemini-3
- https://web.archive.org/web/20251215195306/https://ai.google.dev/gemini-api/docs/pricing
- https://blog.google/products-and-platforms/products/gemini/gemini-3/