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 4.6gpt-oss-120b (batch)
| Claude Opus 4.6 Anthropic | gpt-oss-120b (batch) OpenAI | |
|---|---|---|
| Pricing | ||
| Input $/1M | $5.00 | $0.15 |
| Output $/1M | $25.00 | $0.60 |
| Blended $/1M (3:1) | $10.00 | $0.262 |
| Cache read $/1M | $0.50 | — |
| Cache write $/1M | $6.25 | — |
| Quality indices | ||
| Intelligence Index | 26.4 | 11.6 |
| Coding Index | — | 30.4 |
| Math Index | — | 93.4 |
| Agentic Index | — | 3.7 |
| Speed & latency | ||
| Output speed (tok/s) | 0 | 207 |
| Time to first token | 0s | 0.52s |
| Time to first answer token | 0s | 10.18s |
| Benchmarks (Artificial Analysis) | ||
| MMLU-Pro | — | 80.8% |
| GPQA Diamond | 84.0% | 78.2% |
| Humanity's Last Exam | 19.1% | 19.6% |
| LiveCodeBench | — | 87.8% |
| SciCode | — | 34.0% |
| AIME 2025 | — | 93.4% |
| IFBench | 44.6% | 69.0% |
| AA-LCR (Long Context Reasoning) | 67.0% | 52.0% |
| Terminal-Bench Hard | 48.5% | 23.5% |
| Terminal-Bench 2.1 | — | 26.2% |
| τ²-Bench (Telecom) | 84.8% | 65.8% |
| τ³-Bench Banking | — | 12.8% |
| Design Arena (Elo) | ||
| Website | 1,303 | 980 |
| Web apps | 1,219 | — |
| Full-stack | 1,229 | — |
| Mobile apps | 1,238 | — |
| Android native | 1,216 | — |
| UI components | 1,300 | 942 |
| Data viz | 1,296 | 1,002 |
| SVG | 1,251 | — |
| 3D | 1,303 | 930 |
| Game dev | 1,301 | 1,015 |
| Agentic game dev | 1,215 | — |
| Code categories | 1,304 | 978 |
| ASCII art | 1,273 | — |
| Specs | ||
| Context window | 1M | 131K |
| Max output tokens | 128K | 118K |
| Input modalities | text, image, file | text |
| Tokenizer | Claude | GPT |