Model comparison
Claude Opus 4 vs Gemini 2.5 Pro
Gemini 2.5 Pro is the stronger model overall, scoring 45.0 to 43.1 on the Noometry Index.
Last verified . 55 shared benchmarks.
GoogleSummary
- They share 55 benchmarks with published results for both. Claude Opus 4 scores higher in 4 categories and Gemini 2.5 Pro in 6 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in long context, where Gemini 2.5 Pro leads 59.8 to 39.6.
- The biggest single-benchmark swing is GeoBench: 49% for Claude Opus 4 and 86% for Gemini 2.5 Pro.
- Gemini 2.5 Pro is cheaper at $1.25 / $10 per million input/output tokens, against $15 / $75 for Claude Opus 4.
- Gemini 2.5 Pro accepts more context: 1.05M tokens versus 200K.
Side by side
| Claude Opus 4 | Gemini 2.5 Pro | |
|---|---|---|
| Provider | Anthropic | |
| Noometry Index | 43.1 | 45.0 |
| Released | 2025-05-22 | 2025-03-25 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | 1.05M |
| Max output | 32K | 66K |
| Input $ / M tokens | $15 | $1.25 |
| Output $ / M tokens | $75 | $10 |
| Results tracked | 56 | 78 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 4 leads
Claude Opus 4: 47.2 (#62), Gemini 2.5 Pro: 42.4 (#101)
| Benchmark | Claude Opus 4 | Gemini 2.5 Pro |
|---|---|---|
| SWE-bench Verified | 70.7% | 57.6% |
| SWE-bench Verified (bash only) | 67.6% | 53.6% |
| Aider Polyglot | 72% | 83.1% |
| GSO | 6.9% | 3.9% |
| WeirdML | 43.7% | 54% |
| LMArena Coding | 1442 | 1452 |
| AlgoTune | 1.33 | 1.51 |
| LMArena WebDev | — | 1227 |
| SciCode | — | 42.8% |
| LiveBench Coding | — | 85.9% |
| CadEval | — | 64% |
| ALE-Bench | — | 785.52 |
Agentic & Tool Use Claude Opus 4 leads
Claude Opus 4: 34.8 (#42), Gemini 2.5 Pro: 29.2 (#88)
| Benchmark | Claude Opus 4 | Gemini 2.5 Pro |
|---|---|---|
| DeepResearch Bench | 46.8% | 42.8% |
| LMArena Search | 1127 | 1142 |
| METR Time Horizons | 63.9% | 55.4% |
| Terminal-Bench | — | 32.6% |
| GDPval | — | 23.3% |
| Remote Labor Index | — | 0.8% |
| TheAgentCompany | — | 30.3% |
| τ²-bench Banking | — | 13.7% |
| Cybench | 38% | — |
| BALROG | — | 43.3% |
| Vending-Bench 2 | — | 573.64 |
Reasoning Gemini 2.5 Pro leads
Claude Opus 4: 27.3 (#121), Gemini 2.5 Pro: 28.8 (#99)
| Benchmark | Claude Opus 4 | Gemini 2.5 Pro |
|---|---|---|
| ARC-AGI-2 | 8.6% | 4.9% |
| SimpleBench | 58.8% | 62.4% |
| Kagi LLM Benchmark | 74.3% | 70.3% |
| ARC-AGI-1 | 35.7% | 41% |
| CritPt | 0.3% | 2% |
| EnigmaEval | 5.6% | 5.6% |
| LMArena Hard Prompts | 1399 | 1455 |
| DTBench | 81.6% | 82.4% |
| LMCA | 37.4% | 34.8% |
| Epoch Capabilities Index | 142.67 | 145.32 |
| ForecastBench | 61.1 | 61.3 |
| Chess Puzzles | — | 20% |
| LiveBench Reasoning | — | 89.8% |
| LiveBench Data Analysis | — | 79.9% |
| LiveBench | — | 82.3% |
Math Claude Opus 4 leads
Claude Opus 4: 42.0 (#86), Gemini 2.5 Pro: 32.5 (#213)
| Benchmark | Claude Opus 4 | Gemini 2.5 Pro |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 64.4% | 84.7% |
| Omni-MATH | 61.6% | 41.6% |
| LMArena Math | 1390 | 1450 |
| MATH Level 5 | 85% | 95.9% |
| FrontierMath (Feb 2025 set) | 4.5% | 14.1% |
| FrontierMath Tier 4 (v1) | 4.2% | 4.2% |
| FrontierMath (Tiers 1-3) | — | 24.6% |
| FrontierMath Tier 4 | — | 0% |
| LiveBench Math | — | 90.2% |
Knowledge Gemini 2.5 Pro leads
Claude Opus 4: 44.0 (#88), Gemini 2.5 Pro: 56.0 (#46)
| Benchmark | Claude Opus 4 | Gemini 2.5 Pro |
|---|---|---|
| GPQA Diamond | 76.3% | 85.3% |
| Humanity's Last Exam | 10.7% | 21.6% |
| MMLU-Pro | 87.5% | 86.3% |
| Confabulations | 15.9% | 10.6% |
| Vectara Hallucination Rate | 12% | 7% |
| GPQA (HELM) | 70.8% | 74.9% |
| LMArena Expert | 1386 | 1452 |
Multimodal Gemini 2.5 Pro leads
Claude Opus 4: 31.5 (#106), Gemini 2.5 Pro: 45.2 (#18)
| Benchmark | Claude Opus 4 | Gemini 2.5 Pro |
|---|---|---|
| LMArena Vision | 1192 | 1263 |
| GeoBench | 49% | 86% |
| VPCT | 38% | 48% |
| LMArena Document | — | 1421 |
| SpatialViz-Bench | — | 44.7% |
Multilingual Gemini 2.5 Pro leads
Claude Opus 4: 48.8 (#138), Gemini 2.5 Pro: 55.3 (#31)
| Benchmark | Claude Opus 4 | Gemini 2.5 Pro |
|---|---|---|
| LMArena Non-English | 1362 | 1451 |
| LMArena Chinese | 1386 | 1507 |
| LMArena French | 1372 | 1472 |
| LMArena German | 1391 | 1487 |
| LMArena Japanese | 1331 | 1461 |
| LMArena Korean | 1321 | 1434 |
| LMArena Russian | 1392 | 1461 |
| LMArena Spanish | 1389 | 1473 |
Instruction Following Claude Opus 4 leads
Claude Opus 4: 77.1 (#28), Gemini 2.5 Pro: 75.0 (#75)
| Benchmark | Claude Opus 4 | Gemini 2.5 Pro |
|---|---|---|
| IFEval | 91.8% | 84% |
| LMArena Instruction Following | 1406 | 1437 |
| LiveBench Instruction Following | — | 80.6% |
Long Context Gemini 2.5 Pro leads
Claude Opus 4: 39.6 (#172), Gemini 2.5 Pro: 59.8 (#5)
| Benchmark | Claude Opus 4 | Gemini 2.5 Pro |
|---|---|---|
| Fiction.LiveBench | 61.1% | 91.7% |
| LMArena Longer Query | 1422 | 1449 |
Writing & Preference Gemini 2.5 Pro leads
Claude Opus 4: 61.2 (#89), Gemini 2.5 Pro: 63.7 (#62)
| Benchmark | Claude Opus 4 | Gemini 2.5 Pro |
|---|---|---|
| LMArena Text | 1377 | 1458 |
| LMArena Creative Writing | 1387 | 1454 |
| Short-Story Creative Writing | 83.6% | 83.8% |
| EQ-Bench Creative Writing | 1580 | 1421 |
| WildBench | 85.2% | 85.7% |
| LMArena Multi-Turn | 1396 | 1453 |
| LiveBench Language | — | 67.8% |
Frequently asked questions
Is Claude Opus 4 better than Gemini 2.5 Pro?
Gemini 2.5 Pro is the stronger model overall, scoring 45.0 to 43.1 on the Noometry Index.
Which is cheaper, Claude Opus 4 or Gemini 2.5 Pro?
Gemini 2.5 Pro is cheaper. It lists at $1.25 per million input tokens and $10 per million output tokens; Claude Opus 4 lists at $15 and $75.
Is Claude Opus 4 or Gemini 2.5 Pro better for coding?
Claude Opus 4 scores higher on coding benchmarks: 47.2 versus 42.4 in the Noometry coding category.
Which has the bigger context window?
Gemini 2.5 Pro does, with 1.05M tokens against 200K.
How many benchmarks do Claude Opus 4 and Gemini 2.5 Pro share?
55 benchmarks have published results for both models. Claude Opus 4 has 56 scored results on Noometry and Gemini 2.5 Pro has 78.