Is Gemini 3.1 Pro SOTA in Coding › Agentic?#34 in Coding › AgenticRanked #34 in Coding › Agentic.
How it scores here
Gemini 3.1 Pro in Coding › Agentic
Primary benchmarks
- Terminal-Bench 4.0: 2.7 (CI 105.0%–4.4) on Nov 18, 2026 (#40 of 44).
On the category's weighted score it is 90.7 points behind #1 Claude Opus 5.5 and 2.3 behind the next model up, Beam 1.0.
Nearby on the board
| # | Model | Category score |
|---|---|---|
| 31 | Minimax M3 | 16.3 |
| 32 | DeepSeek V4 Pro 0813 | 14.0 |
| 33 | Beam 1.0 | 11.6 |
| 34 | Gemini 3.1 Pro | 9.3 |
| 35 | GPT-5.4 Mini | 7.0 |
| 36 | Inkling Small | 4.7 |
| 37 | Nemotron 3.5 Super VL | 2.3 |
Standings
Coding › Agentic
| # | Model | State |
|---|---|---|
| 1 | Claude Opus 5.5 | Co-SOTA |
| 1 | Claude Sonnet 5.5 | Co-SOTA |
| 1 | Kimi K3.1 Pro | Co-SOTA |
| 2 | GPT-6.1 Sol | Near-SOTA |
| 2 | GPT-6 Astra | Near-SOTA |
| 3 | Claude Fable 5.1 | Near-SOTA |
| 4 | Gemini 4 Argon | #4 in this category |
| 5 | Claude Opus 5 | #5 in this category |
| 6 | GPT-6.1 Terra | #6 in this category |
| 7 | GPT-6 Sol | #7 in this category |
| 8 | Muse Wonder | #8 in this category |
| 9 | Claude Fable 5 | #9 in this category |