COMPARE
Compare AI models
Performance and cost, for your workload.
Prices & metadata · 2026-10-25Benchmarks · 2026-10-23ECI · 2026-10-24
GPQA Diamond3 models · 1M input + 250K output tokens
| Compare modelsBest available values highlighted |  |  |  |
|---|---|---|---|
| Price | |||
| Input / 1M tokens | $2.00 | $10.00 | $0.15 |
| Output / 1M tokens | $10.00 | $50.00 | $0.60 |
| Your workload | $4.50 | $22.50 | $0.30 |
| Benchmarks | |||
| GPQA Diamond | 95.6% | 95.8% | 93.2% |
| FrontierMath tiers 1–3 · v2 | 88.8% | 93.7% | 71.2% |
| SimpleQA Verified | 46.5% | 75.6% | 42.5% |
| OTIS Mock AIME 2024–2025 | 100.0% | 100.0% | 99.4% |
| Capacity | |||
| Context window | 1M | 1.05M | 1.05M |
| Maximum output | 128K | 128K | 393.22K |
| ECI estimate90% interval | 166.2163.1–171.3 | 168.8165.4–174.5 | — |
| Capabilities | |||
| Tool calling | + Listed | + Listed | + Listed |
| Reasoning controls | + Listed | + Listed | + Listed |
| Image input | + Listed | + Listed | + Listed |
| Audio input | − Not listed | − Not listed | − Not listed |
| Image output | − Not listed | − Not listed | − Not listed |
| Input types | text, image, file | file, image, text | text, image |
| Output types | text | text | text |
Highlights compare available values in each row, including ties. Capacity is not quality; overlapping ECI intervals do not establish a winner. − means not listed, not confirmed unsupported.
Listed prices; no caching or batch discounts. Each benchmark is separate, with differing evaluation dates and compute budgets.
Lower is better for price. Larger context windows describe capacity, not model quality. Missing values remain unavailable.
