Model comparison
Claude Sonnet 4 vs Gemini 3.1 Pro Preview
Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 40.8 on the Noometry Index.
Last verified . 40 shared benchmarks.
GoogleSummary
- They share 40 benchmarks with published results for both. Claude Sonnet 4 scores higher in 2 categories and Gemini 3.1 Pro Preview in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.1 Pro Preview leads 71.7 to 22.9.
- The biggest single-benchmark swing is ARC-AGI-2: 5.9% for Claude Sonnet 4 and 77.1% for Gemini 3.1 Pro Preview.
- Gemini 3.1 Pro Preview is cheaper at $2 / $12 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.
- Gemini 3.1 Pro Preview accepts more context: 1.05M tokens versus 200K.
Side by side
| Claude Sonnet 4 | Gemini 3.1 Pro Preview | |
|---|---|---|
| Provider | Anthropic | |
| Noometry Index | 40.8 | 56.7 |
| Released | 2025-05-22 | 2026-02-19 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | 1.05M |
| Max output | 64K | 66K |
| Input $ / M tokens | $3 | $2 |
| Output $ / M tokens | $15 | $12 |
| Results tracked | 58 | 71 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Claude Sonnet 4: 43.5 (#88), Gemini 3.1 Pro Preview: 42.5 (#99)
| Benchmark | Claude Sonnet 4 | Gemini 3.1 Pro Preview |
|---|---|---|
| SciCode | 40% | 58.9% |
| GSO | 4.9% | 22.6% |
| WeirdML | 46.1% | 72.1% |
| LMArena Coding | 1414 | 1484 |
| ALE-Bench | 655.35 | 1,161 |
| SWE-bench Verified | — | 75.6% |
| DeepSWE | — | 11.7% |
| SWE-bench Verified (bash only) | 64.9% | — |
| Aider Polyglot | 61.3% | — |
| LMArena WebDev | — | 1447 |
| MirrorCode | — | 8.9% |
| AlgoTune | — | 2.02 |
Agentic & Tool Use Too close to call
Claude Sonnet 4: 38.5 (#31), Gemini 3.1 Pro Preview: 37.7 (#34)
| Benchmark | Claude Sonnet 4 | Gemini 3.1 Pro Preview |
|---|---|---|
| DeepResearch Bench | 46.6% | 47.8% |
| METR Time Horizons | 62% | 77% |
| Terminal-Bench | — | 80.2% |
| APEX-Agents | — | 35.3% |
| TheAgentCompany | 33.1% | — |
| τ²-bench Banking | — | 26% |
| Cybench | 35% | — |
| OSWorld | 43.9% | — |
| PostTrainBench | — | 22% |
| BALROG | — | 57% |
| ExploitBench | — | 26.1% |
| GBAEval | — | 0.8% |
| GDP.pdf | — | 17% |
| LMArena Search | — | 1211 |
| Vending-Bench 2 | — | 3,774 |
Reasoning Gemini 3.1 Pro Preview leads
Claude Sonnet 4: 22.9 (#187), Gemini 3.1 Pro Preview: 71.7 (#12)
| Benchmark | Claude Sonnet 4 | Gemini 3.1 Pro Preview |
|---|---|---|
| ARC-AGI-2 | 5.9% | 77.1% |
| SimpleBench | 45.5% | 79.6% |
| ARC-AGI-1 | 40% | 98% |
| CritPt | 0.3% | 17.7% |
| EnigmaEval | 3.1% | 36.8% |
| LMArena Hard Prompts | 1372 | 1485 |
| DTBench | 77.1% | 97.1% |
| LMCA | 29% | 53.8% |
| Epoch Capabilities Index | 141.69 | 154.77 |
| ForecastBench | 60.2 | 59 |
| Kagi LLM Benchmark | 73% | — |
| NYT Connections (extended) | — | 97.4% |
| Chess Puzzles | — | 55% |
| Thematic Generalization | — | 79.4% |
| EBR-Bench | — | 14.3% |
| Mystery Game Puzzles | — | 34% |
Math Gemini 3.1 Pro Preview leads
Claude Sonnet 4: 43.3 (#80), Gemini 3.1 Pro Preview: 62.1 (#34)
| Benchmark | Claude Sonnet 4 | Gemini 3.1 Pro Preview |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 95.6% |
| LMArena Math | 1375 | 1485 |
| FrontierMath (Feb 2025 set) | 4.1% | 36.9% |
| FrontierMath Tier 4 (v1) | 0% | 16.7% |
| FrontierMath (Tiers 1-3) | — | 59.6% |
| FrontierMath Tier 4 | — | 26.8% |
| MathArena Final-Answer Competitions | — | 86.5% |
| ProofBench | — | 26% |
| Omni-MATH | 60.2% | — |
| MATH Level 5 | 84.4% | — |
Knowledge Gemini 3.1 Pro Preview leads
Claude Sonnet 4: 41.8 (#108), Gemini 3.1 Pro Preview: 71.8 (#3)
| Benchmark | Claude Sonnet 4 | Gemini 3.1 Pro Preview |
|---|---|---|
| GPQA Diamond | 79.2% | 94.4% |
| Humanity's Last Exam | 7.8% | 46.4% |
| Vectara Hallucination Rate | 10.3% | 10.4% |
| LMArena Expert | 1372 | 1485 |
| SimpleQA Verified | — | 73.5% |
| MMLU-Pro | 84.3% | — |
| Confabulations | 13.2% | — |
| GPQA (HELM) | 70.6% | — |
Multimodal Gemini 3.1 Pro Preview leads
Claude Sonnet 4: 26.2 (#121), Gemini 3.1 Pro Preview: 37.9 (#69)
| Benchmark | Claude Sonnet 4 | Gemini 3.1 Pro Preview |
|---|---|---|
| LMArena Vision | 1191 | 1296 |
| GeoBench | 37% | — |
| VPCT | 34% | — |
| Blueprint-Bench 2 | — | 26.5% |
| Furniture Assembly | — | 26.7% |
| LMArena Document | — | 1444 |
| MindCube | 44.8% | — |
Multilingual Gemini 3.1 Pro Preview leads
Claude Sonnet 4: 46.7 (#156), Gemini 3.1 Pro Preview: 57.0 (#12)
| Benchmark | Claude Sonnet 4 | Gemini 3.1 Pro Preview |
|---|---|---|
| LMArena Non-English | 1333 | 1477 |
| LMArena Chinese | 1350 | 1529 |
| LMArena French | 1363 | 1487 |
| LMArena German | 1331 | 1491 |
| LMArena Japanese | 1302 | 1493 |
| LMArena Korean | 1291 | 1455 |
| LMArena Russian | 1355 | 1498 |
| LMArena Spanish | 1357 | 1479 |
Instruction Following Gemini 3.1 Pro Preview leads
Claude Sonnet 4: 71.7 (#145), Gemini 3.1 Pro Preview: 77.0 (#32)
| Benchmark | Claude Sonnet 4 | Gemini 3.1 Pro Preview |
|---|---|---|
| LMArena Instruction Following | 1376 | 1466 |
| IFEval | 84% | — |
Long Context Gemini 3.1 Pro Preview leads
Claude Sonnet 4: 33.7 (#259), Gemini 3.1 Pro Preview: 47.4 (#18)
| Benchmark | Claude Sonnet 4 | Gemini 3.1 Pro Preview |
|---|---|---|
| LMArena Longer Query | 1398 | 1483 |
| Fiction.LiveBench | 46.9% | — |
| CL-bench | — | 20.8% |
| CL-bench Life | — | 16.9% |
Writing & Preference Gemini 3.1 Pro Preview leads
Claude Sonnet 4: 57.1 (#132), Gemini 3.1 Pro Preview: 66.1 (#37)
| Benchmark | Claude Sonnet 4 | Gemini 3.1 Pro Preview |
|---|---|---|
| LMArena Text | 1351 | 1481 |
| LMArena Creative Writing | 1345 | 1482 |
| EQ-Bench Creative Writing | 1483 | 1491 |
| LMArena Multi-Turn | 1376 | 1488 |
| Short-Story Creative Writing | 81.4% | — |
| WildBench | 83.8% | — |
| EQ-Bench 4 | — | 1142 |
Frequently asked questions
Is Claude Sonnet 4 better than Gemini 3.1 Pro Preview?
Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 40.8 on the Noometry Index.
Which is cheaper, Claude Sonnet 4 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 Sonnet 4 lists at $3 and $15.
Is Claude Sonnet 4 or Gemini 3.1 Pro Preview better for coding?
They score almost the same on coding (43.5 vs 42.5); test both on your own repository before choosing.
Which has the bigger context window?
Gemini 3.1 Pro Preview does, with 1.05M tokens against 200K.
How many benchmarks do Claude Sonnet 4 and Gemini 3.1 Pro Preview share?
40 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and Gemini 3.1 Pro Preview has 71.