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 Small 4GPT-5.6 Sol
| Mistral Small 4 Mistral | GPT-5.6 Sol OpenAI | |
|---|---|---|
| Pricing | ||
| Input $/1M | $0.15 | $5.00 |
| Output $/1M | $0.60 | $30.00 |
| Blended $/1M (3:1) | $0.262 | $11.25 |
| Cache read $/1M | $0.015 | $0.50 |
| Cache write $/1M | — | $6.25 |
| Quality indices | ||
| Intelligence Index | 19.6 | 58.9 |
| Coding Index | 26.6 | 77.4 |
| Math Index | — | — |
| Agentic Index | 4.7 | 54 |
| Speed & latency | ||
| Output speed (tok/s) | 175 | 67 |
| Time to first token | 0.53s | 120.38s |
| Time to first answer token | 11.93s | 120.38s |
| Benchmarks (Artificial Analysis) | ||
| GPQA Diamond | 76.9% | 94.1% |
| Humanity's Last Exam | 9.5% | 47.2% |
| SciCode | 38.0% | 56.1% |
| IFBench | 48.2% | 72.7% |
| AA-LCR (Long Context Reasoning) | 44.7% | 73.7% |
| Terminal-Bench Hard | 17.4% | 65.9% |
| Terminal-Bench 2.1 | 21.0% | 88.0% |
| τ²-Bench (Telecom) | 41.2% | 85.1% |
| τ³-Bench Banking | 5.2% | 33.0% |
| Specs | ||
| Context window | 262K | 1.05M |
| Max output tokens | — | 128K |
| Input modalities | text, image | file, image, text |
| Tokenizer | Mistral | GPT |