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 4.5Grok 4.20
| Claude Haiku 4.5 Anthropic | Grok 4.20 xAI | |
|---|---|---|
| Pricing | ||
| Input $/1M | $1.00 | $1.25 |
| Output $/1M | $5.00 | $2.50 |
| Blended $/1M (3:1) | $2.00 | $1.56 |
| Cache read $/1M | $0.10 | $0.20 |
| Cache write $/1M | $1.25 | — |
| Quality indices | ||
| Intelligence Index | 29.6 | 37 |
| Coding Index | 43.9 | — |
| Math Index | — | — |
| Agentic Index | 16.4 | — |
| Speed & latency | ||
| Output speed (tok/s) | — | 230 |
| Time to first token | — | 16.10s |
| Time to first answer token | — | 16.10s |
| Benchmarks (Artificial Analysis) | ||
| GPQA Diamond | — | 91.1% |
| Humanity's Last Exam | — | 32.2% |
| SciCode | — | 45.6% |
| IFBench | — | 81.2% |
| AA-LCR (Long Context Reasoning) | — | 58.0% |
| Terminal-Bench Hard | — | 37.9% |
| τ²-Bench (Telecom) | — | 93.0% |
| Design Arena (Elo) | ||
| Website | 1,149 | 1,255 |
| Web apps | — | 1,200 |
| Full-stack | — | 1,111 |
| Mobile apps | — | 1,181 |
| Android native | — | 1,154 |
| UI components | 1,142 | 1,245 |
| Data viz | 1,160 | 1,241 |
| SVG | 1,072 | 1,211 |
| 3D | 1,131 | 1,249 |
| Game dev | 1,155 | 1,255 |
| Godot game dev | — | 1,089 |
| HTML slides | — | 1,180 |
| Code categories | 1,148 | 1,252 |
| ASCII art | 1,182 | 1,219 |
| Specs | ||
| Context window | 200K | 2M |
| Max output tokens | 64K | — |
| Input modalities | text, image, file | text, image, file |
| Tokenizer | Claude | Grok |