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.
GPT-5.6 SolQwen3 VL 32B Instruct
| GPT-5.6 Sol OpenAI | Qwen3 VL 32B Instruct Alibaba (Qwen) | |
|---|---|---|
| Pricing | ||
| Input $/1M | $5.00 | $0.104 |
| Output $/1M | $30.00 | $0.416 |
| Blended $/1M (3:1) | $11.25 | $0.182 |
| Cache read $/1M | $0.50 | — |
| Cache write $/1M | $6.25 | — |
| Quality indices | ||
| Intelligence Index | 58.9 | 11.1 |
| Coding Index | 77.4 | — |
| Math Index | — | 68.3 |
| Agentic Index | 54 | — |
| Speed & latency | ||
| Output speed (tok/s) | 67 | 73 |
| Time to first token | 120.38s | 1.12s |
| Time to first answer token | 120.38s | 1.12s |
| Benchmarks (Artificial Analysis) | ||
| MMLU-Pro | — | 79.1% |
| GPQA Diamond | 94.1% | 67.1% |
| Humanity's Last Exam | 47.2% | 6.3% |
| LiveCodeBench | — | 51.4% |
| SciCode | 56.1% | 30.1% |
| AIME 2025 | — | 68.3% |
| IFBench | 72.7% | 39.2% |
| AA-LCR (Long Context Reasoning) | 73.7% | 31.3% |
| Terminal-Bench Hard | 65.9% | 8.3% |
| Terminal-Bench 2.1 | 88.0% | — |
| τ²-Bench (Telecom) | 85.1% | 29.2% |
| τ³-Bench Banking | 33.0% | — |
| Specs | ||
| Context window | 1.05M | 131K |
| Max output tokens | 128K | 33K |
| Input modalities | file, image, text | text, image |
| Tokenizer | GPT | Qwen |