# Claude Opus 4.5 vs Gemini 3.1 Pro Preview

> Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 50.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-opus-4-5-vs-gemini-3-1-pro-preview
- Last updated: 2026-10-12
- Shared benchmarks: 56

## Summary

- They share 56 benchmarks with published results for both. Claude Opus 4.5 scores higher in 4 categories and Gemini 3.1 Pro Preview in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.1 Pro Preview leads 71.7 to 42.6.
- The biggest single-benchmark swing is NYT Connections (extended): 52.5% for Claude Opus 4.5 and 97.4% for Gemini 3.1 Pro Preview.
- Gemini 3.1 Pro Preview is cheaper at $2 / $12 per million input/output tokens, against $5 / $25 for Claude Opus 4.5.
- Gemini 3.1 Pro Preview accepts more context: 1.05M tokens versus 200K.

## Snapshot

| | Claude Opus 4.5 | Gemini 3.1 Pro Preview |
|---|---|---|
| Provider | Anthropic | Google |
| Noometry Index | 50.5 | 56.7 |
| Rank | 47 | 23 |
| Context | 200K | 1.05M |
| Input $/M | $5 | $2 |
| Output $/M | $25 | $12 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Opus 4.5: 54.8 (#27)
- Gemini 3.1 Pro Preview: 42.5 (#99)

| Benchmark | Claude Opus 4.5 | Gemini 3.1 Pro Preview |
|---|---|---|
| SWE-bench Verified | 76.7% | 75.6% |
| LMArena WebDev | 1494 | 1447 |
| GSO | 26.5% | 22.6% |
| WeirdML | 63.7% | 72.1% |
| LMArena Coding | 1504 | 1484 |
| ALE-Bench | 1,025 | 1,161 |
| AlgoTune | 1.77 | 2.02 |
| DeepSWE | — | 11.7% |
| SWE-bench Verified (bash only) | 76.8% | — |
| SWE-bench Multilingual | 70.7% | — |
| SciCode | — | 58.9% |
| MirrorCode | — | 8.9% |

## Agentic & Tool Use

- Claude Opus 4.5: 47.3 (#12)
- Gemini 3.1 Pro Preview: 37.7 (#34)

| Benchmark | Claude Opus 4.5 | Gemini 3.1 Pro Preview |
|---|---|---|
| Terminal-Bench | 63.1% | 80.2% |
| τ²-bench Banking | 24.7% | 26% |
| DeepResearch Bench | 54.8% | 47.8% |
| BALROG | 43.5% | 57% |
| LMArena Search | 1180 | 1211 |
| METR Time Horizons | 75% | 77% |
| Vending-Bench 2 | 4,967 | 3,774 |
| APEX-Agents | — | 35.3% |
| Berkeley Function Calling Leaderboard | 77.5% | — |
| GDPval | 45.5% | — |
| Remote Labor Index | 3.8% | — |
| τ²-bench Airline | 84% | — |
| τ²-bench Retail | 79.6% | — |
| τ²-bench Telecom | 92.3% | — |
| Cybench | 82% | — |
| OSWorld | 66.3% | — |
| PostTrainBench | — | 22% |
| ExploitBench | — | 26.1% |
| GBAEval | — | 0.8% |
| GDP.pdf | — | 17% |

## Reasoning

- Claude Opus 4.5: 42.6 (#51)
- Gemini 3.1 Pro Preview: 71.7 (#12)

| Benchmark | Claude Opus 4.5 | Gemini 3.1 Pro Preview |
|---|---|---|
| ARC-AGI-2 | 37.6% | 77.1% |
| SimpleBench | 62% | 79.6% |
| NYT Connections (extended) | 52.5% | 97.4% |
| ARC-AGI-1 | 80% | 98% |
| Chess Puzzles | 12% | 55% |
| EnigmaEval | 11.9% | 36.8% |
| EBR-Bench | 14.3% | 14.3% |
| LMArena Hard Prompts | 1476 | 1485 |
| Mystery Game Puzzles | 22% | 34% |
| DTBench | 89.9% | 97.1% |
| LMCA | 44.5% | 53.8% |
| Epoch Capabilities Index | 150.09 | 154.77 |
| ForecastBench | 60.7 | 59 |
| Kagi LLM Benchmark | 80.2% | — |
| CritPt | — | 17.7% |
| Thematic Generalization | — | 79.4% |

## Math

- Claude Opus 4.5: 38.6 (#132)
- Gemini 3.1 Pro Preview: 62.1 (#34)

| Benchmark | Claude Opus 4.5 | Gemini 3.1 Pro Preview |
|---|---|---|
| FrontierMath (Tiers 1-3) | 34.4% | 59.6% |
| FrontierMath Tier 4 | 4.9% | 26.8% |
| OTIS Mock AIME 2024-2025 | 86.1% | 95.6% |
| ProofBench | 36% | 26% |
| LMArena Math | 1463 | 1485 |
| FrontierMath (Feb 2025 set) | 20.7% | 36.9% |
| FrontierMath Tier 4 (v1) | 4.2% | 16.7% |
| MathArena Final-Answer Competitions | — | 86.5% |

## Knowledge

- Claude Opus 4.5: 56.5 (#44)
- Gemini 3.1 Pro Preview: 71.8 (#3)

| Benchmark | Claude Opus 4.5 | Gemini 3.1 Pro Preview |
|---|---|---|
| GPQA Diamond | 86% | 94.4% |
| Humanity's Last Exam | 25.2% | 46.4% |
| SimpleQA Verified | 45.7% | 73.5% |
| Vectara Hallucination Rate | 10.9% | 10.4% |
| LMArena Expert | 1487 | 1485 |

## Multimodal

- Claude Opus 4.5: 31.4 (#107)
- Gemini 3.1 Pro Preview: 37.9 (#69)

| Benchmark | Claude Opus 4.5 | Gemini 3.1 Pro Preview |
|---|---|---|
| Furniture Assembly | 28.3% | 26.7% |
| LMArena Document | 1462 | 1444 |
| LMArena Vision | — | 1296 |
| GeoBench | 75% | — |
| VPCT | 40% | — |
| Blueprint-Bench 2 | — | 26.5% |

## Multilingual

- Claude Opus 4.5: 54.3 (#47)
- Gemini 3.1 Pro Preview: 57.0 (#12)

| Benchmark | Claude Opus 4.5 | Gemini 3.1 Pro Preview |
|---|---|---|
| LMArena Non-English | 1438 | 1477 |
| LMArena Chinese | 1470 | 1529 |
| LMArena French | 1471 | 1487 |
| LMArena German | 1449 | 1491 |
| LMArena Japanese | 1416 | 1493 |
| LMArena Korean | 1424 | 1455 |
| LMArena Russian | 1447 | 1498 |
| LMArena Spanish | 1458 | 1479 |

## Instruction Following

- Claude Opus 4.5: 77.5 (#19)
- Gemini 3.1 Pro Preview: 77.0 (#32)

| Benchmark | Claude Opus 4.5 | Gemini 3.1 Pro Preview |
|---|---|---|
| LMArena Instruction Following | 1478 | 1466 |

## Long Context

- Claude Opus 4.5: 46.5 (#22)
- Gemini 3.1 Pro Preview: 47.4 (#18)

| Benchmark | Claude Opus 4.5 | Gemini 3.1 Pro Preview |
|---|---|---|
| CL-bench | 21.1% | 20.8% |
| LMArena Longer Query | 1480 | 1483 |
| CL-bench Life | — | 16.9% |

## Writing & Preference

- Claude Opus 4.5: 68.1 (#28)
- Gemini 3.1 Pro Preview: 66.1 (#37)

| Benchmark | Claude Opus 4.5 | Gemini 3.1 Pro Preview |
|---|---|---|
| LMArena Text | 1451 | 1481 |
| LMArena Creative Writing | 1445 | 1482 |
| EQ-Bench Creative Writing | 1687 | 1491 |
| LMArena Multi-Turn | 1466 | 1488 |
| EQ-Bench 4 | — | 1142 |

## FAQ

### Is Claude Opus 4.5 better than Gemini 3.1 Pro Preview?

Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 50.5 on the Noometry Index.

### Which is cheaper, Claude Opus 4.5 or Gemini 3.1 Pro Preview?

Gemini 3.1 Pro Preview is cheaper. It lists at $2 per million input tokens and $12 per million output tokens; Claude Opus 4.5 lists at $5 and $25.

### Is Claude Opus 4.5 or Gemini 3.1 Pro Preview better for coding?

Claude Opus 4.5 scores higher on coding benchmarks: 54.8 versus 42.5 in the Noometry coding category.

### Which has the bigger context window?

Gemini 3.1 Pro Preview does, with 1.05M tokens against 200K.

### How many benchmarks do Claude Opus 4.5 and Gemini 3.1 Pro Preview share?

56 benchmarks have published results for both models. Claude Opus 4.5 has 69 scored results on Noometry and Gemini 3.1 Pro Preview has 71.