Compare models
Pick up to 4 models for a side-by-side spec sheet. Hover any dotted metric name for what the test measures. Best value in each row is highlighted.
Claude Opus 4.6Gemini 2.5 Pro
| Claude Opus 4.6 Anthropic | Gemini 2.5 Pro | |
|---|---|---|
| Pricing | ||
| Input $/1M | $5.00 | $1.25 |
| Output $/1M | $25.00 | $10.00 |
| Blended $/1M (3:1) | $10.00 | $3.44 |
| Cache read $/1M | $0.50 | $0.125 |
| Cache write $/1M | $6.25 | $0.375 |
| Quality indices | ||
| Intelligence Index | 37.8 | 25.8 |
| Coding Index | — | 33.3 |
| Math Index | — | 87.7 |
| Agentic Index | — | 7.1 |
| Speed & latency | ||
| Output speed (tok/s) | 54 | 144 |
| Time to first token | 1.89s | 20.10s |
| Time to first answer token | 1.89s | 20.10s |
| Benchmarks (Artificial Analysis) | ||
| MMLU-Pro | — | 86.2% |
| GPQA Diamond | 84.0% | 84.4% |
| Humanity's Last Exam | 18.6% | 21.1% |
| LiveCodeBench | — | 80.1% |
| SciCode | 45.7% | 42.8% |
| MATH-500 | — | 96.7% |
| AIME | — | 88.7% |
| AIME 2025 | — | 87.7% |
| IFBench | 44.6% | 48.7% |
| AA-LCR (Long Context Reasoning) | 58.3% | 66.0% |
| Terminal-Bench Hard | 48.5% | 26.5% |
| Terminal-Bench 2.1 | — | 28.5% |
| τ²-Bench (Telecom) | 84.8% | 54.1% |
| τ³-Bench Banking | — | 9.3% |
| Design Arena (Elo) | ||
| Website | 1,323 | 1,194 |
| Web apps | 1,261 | — |
| Full-stack | 1,279 | — |
| Mobile apps | 1,280 | — |
| Android native | 1,195 | — |
| UI components | 1,336 | 1,186 |
| Data viz | 1,322 | 1,283 |
| SVG | 1,274 | — |
| 3D | 1,335 | 1,141 |
| Game dev | 1,337 | 1,171 |
| Code categories | 1,327 | 1,187 |
| ASCII art | 1,298 | — |
| Specs | ||
| Context window | 1M | 1.05M |
| Max output tokens | 128K | 66K |
| Input modalities | text, image, file | text, image, file, audio, video |
| Tokenizer | Claude | Gemini |