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.
Mistral Medium 3.5 (batch)GPT-5.6 Sol
| Mistral Medium 3.5 (batch) Mistral | GPT-5.6 Sol OpenAI | |
|---|---|---|
| Pricing | ||
| Input $/1M | $0.75 | $2.00 |
| Output $/1M | $3.75 | $10.00 |
| Blended $/1M (3:1) | $1.50 | $4.00 |
| Cache read $/1M | — | $0.20 |
| Cache write $/1M | — | $2.50 |
| Quality indices | ||
| Intelligence Index | 14.2 | 47 |
| Coding Index | 46.9 | 77.4 |
| Math Index | — | — |
| Agentic Index | 7 | 50.2 |
| Speed & latency | ||
| Output speed (tok/s) | 152 | 65 |
| Time to first token | 0.65s | 64.46s |
| Time to first answer token | 13.83s | 64.46s |
| Benchmarks (Artificial Analysis) | ||
| GPQA Diamond | 74.8% | 94.1% |
| Humanity's Last Exam | 13.8% | 49.5% |
| SciCode | 40.2% | 57.1% |
| IFBench | 68.8% | 72.7% |
| AA-LCR (Long Context Reasoning) | 69.3% | 84.0% |
| Terminal-Bench Hard | 33.3% | 65.9% |
| Terminal-Bench 2.1 | 50.6% | 88.0% |
| τ²-Bench (Telecom) | 94.2% | 85.1% |
| τ³-Bench Banking | 15.1% | 44.3% |
| Specs | ||
| Context window | 262K | 1.05M |
| Max output tokens | 210K | 128K |
| Input modalities | text, image, file | file, image, text |
| Tokenizer | Mistral | GPT |