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.
Claude Haiku 4.5Qwen3 235B A22B Thinking 2507
| Claude Haiku 4.5 Anthropic | Qwen3 235B A22B Thinking 2507 Alibaba (Qwen) | |
|---|---|---|
| Pricing | ||
| Input $/1M | $1.00 | $0.23 |
| Output $/1M | $5.00 | $2.30 |
| Blended $/1M (3:1) | $2.00 | $0.747 |
| Cache read $/1M | $0.10 | — |
| Cache write $/1M | $1.25 | — |
| Quality indices | ||
| Intelligence Index | 16.9 | 12.7 |
| Coding Index | 43.9 | 22.1 |
| Math Index | — | — |
| Agentic Index | 8 | 1.3 |
| Speed & latency | ||
| Output speed (tok/s) | — | — |
| Time to first token | — | — |
| Time to first answer token | — | — |
| Design Arena (Elo) | ||
| Website | 1,135 | 1,065 |
| UI components | 1,115 | 960 |
| Data viz | 1,139 | 964 |
| SVG | 1,049 | — |
| 3D | 1,101 | 1,026 |
| Game dev | 1,120 | 980 |
| Code categories | 1,130 | 1,048 |
| ASCII art | 1,159 | — |
| Specs | ||
| Context window | 200K | 131K |
| Max output tokens | 64K | 118K |
| Input modalities | text, image, file | text |
| Tokenizer | Claude | Qwen3 |