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 235B A22B Instruct
| GPT-5.6 Sol OpenAI | Qwen3 VL 235B A22B Instruct Alibaba (Qwen) | |
|---|---|---|
| Pricing | ||
| Input $/1M | $2.00 | $0.21 |
| Output $/1M | $10.00 | $1.90 |
| Blended $/1M (3:1) | $4.00 | $0.632 |
| Cache read $/1M | $0.20 | $0.10 |
| Cache write $/1M | $2.50 | — |
| Quality indices | ||
| Intelligence Index | 47 | 9.9 |
| Coding Index | 77.4 | — |
| Math Index | — | 70.7 |
| Agentic Index | 50.2 | — |
| Speed & latency | ||
| Output speed (tok/s) | 65 | 0 |
| Time to first token | 64.46s | 0s |
| Time to first answer token | 64.46s | 0s |
| Benchmarks (Artificial Analysis) | ||
| MMLU-Pro | — | 82.3% |
| GPQA Diamond | 94.1% | 71.2% |
| Humanity's Last Exam | 49.5% | 6.6% |
| LiveCodeBench | — | 59.4% |
| SciCode | 57.1% | — |
| AIME 2025 | — | 70.7% |
| IFBench | 72.7% | 42.7% |
| AA-LCR (Long Context Reasoning) | 84.0% | 32.7% |
| Terminal-Bench Hard | 65.9% | 6.8% |
| Terminal-Bench 2.1 | 88.0% | — |
| τ²-Bench (Telecom) | 85.1% | 35.1% |
| τ³-Bench Banking | 44.3% | — |
| Specs | ||
| Context window | 1.05M | 262K |
| Max output tokens | 128K | 33K |
| Input modalities | file, image, text | text, image |
| Tokenizer | GPT | Qwen3 |