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 Opus 4.6Qwen3.7 Max
| Claude Opus 4.6 Anthropic | Qwen3.7 Max Alibaba (Qwen) | |
|---|---|---|
| Pricing | ||
| Input $/1M | $5.00 | $1.48 |
| Output $/1M | $25.00 | $4.42 |
| Blended $/1M (3:1) | $10.00 | $2.21 |
| Cache read $/1M | $0.50 | $0.295 |
| Cache write $/1M | $6.25 | $1.84 |
| Quality indices | ||
| Intelligence Index | 37.8 | 46 |
| Coding Index | — | 66 |
| Math Index | — | — |
| Agentic Index | — | 30.6 |
| Speed & latency | ||
| Output speed (tok/s) | 54 | 207 |
| Time to first token | 1.89s | 1.55s |
| Time to first answer token | 1.89s | 13.16s |
| Benchmarks (Artificial Analysis) | ||
| GPQA Diamond | 84.0% | 92.3% |
| Humanity's Last Exam | 18.6% | 38.1% |
| SciCode | 45.7% | 48.8% |
| IFBench | 44.6% | 80.5% |
| AA-LCR (Long Context Reasoning) | 58.3% | 69.0% |
| Terminal-Bench Hard | 48.5% | 50.8% |
| Terminal-Bench 2.1 | — | 74.5% |
| τ²-Bench (Telecom) | 84.8% | 94.7% |
| τ³-Bench Banking | — | 10.9% |
| Design Arena (Elo) | ||
| Website | 1,323 | 1,291 |
| Web apps | 1,261 | 1,253 |
| Full-stack | 1,279 | 1,225 |
| Mobile apps | 1,280 | 1,208 |
| Android native | 1,195 | 1,182 |
| UI components | 1,336 | 1,313 |
| Data viz | 1,322 | 1,304 |
| SVG | 1,274 | 1,267 |
| 3D | 1,335 | 1,310 |
| Game dev | 1,337 | 1,315 |
| Agentic game dev | — | 1,180 |
| Godot game dev | — | 1,248 |
| HTML slides | — | 1,191 |
| Python→PPTX slides | — | 1,229 |
| Code categories | 1,327 | 1,299 |
| ASCII art | 1,298 | 1,261 |
| Specs | ||
| Context window | 1M | 1M |
| Max output tokens | 128K | 66K |
| Input modalities | text, image, file | text |
| Tokenizer | Claude | Qwen |