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 4.5Gemini 3.1 Pro Preview
| Claude Haiku 4.5 Anthropic | Gemini 3.1 Pro Preview | |
|---|---|---|
| Pricing | ||
| Input $/1M | $1.00 | $2.00 |
| Output $/1M | $5.00 | $12.00 |
| Blended $/1M (3:1) | $2.00 | $4.50 |
| Cache read $/1M | $0.10 | $0.20 |
| Cache write $/1M | $1.25 | $0.375 |
| Quality indices | ||
| Intelligence Index | 29.6 | 46.5 |
| Coding Index | 43.9 | 68.8 |
| Math Index | — | — |
| Agentic Index | 16.4 | 21.4 |
| Speed & latency | ||
| Output speed (tok/s) | — | 135 |
| Time to first token | — | 19.70s |
| Time to first answer token | — | 19.70s |
| Benchmarks (Artificial Analysis) | ||
| GPQA Diamond | — | 94.1% |
| Humanity's Last Exam | — | 44.7% |
| SciCode | — | 58.9% |
| IFBench | — | 77.1% |
| AA-LCR (Long Context Reasoning) | — | 72.7% |
| Terminal-Bench Hard | — | 53.8% |
| Terminal-Bench 2.1 | — | 73.8% |
| τ²-Bench (Telecom) | — | 95.6% |
| τ³-Bench Banking | — | 16.5% |
| Design Arena (Elo) | ||
| Website | 1,149 | 1,278 |
| Web apps | — | 1,182 |
| Full-stack | — | 1,127 |
| Mobile apps | — | 1,167 |
| Android native | — | 1,038 |
| UI components | 1,142 | 1,309 |
| Data viz | 1,160 | 1,262 |
| SVG | 1,072 | 1,334 |
| 3D | 1,131 | 1,291 |
| Game dev | 1,155 | 1,258 |
| Agentic game dev | — | 1,128 |
| Godot game dev | — | 1,159 |
| HTML slides | — | 1,202 |
| PPTX slides | — | 1,110 |
| Python→PPTX slides | — | 1,109 |
| Agentic slides | — | 1,112 |
| Agentic HTML slides | — | 1,226 |
| Agentic slides (HTML) | — | 1,219 |
| Agentic slides (Python PPTX) | — | 1,107 |
| Code categories | 1,148 | 1,274 |
| ASCII art | 1,182 | 1,312 |
| Specs | ||
| Context window | 200K | 1.05M |
| Max output tokens | 64K | 66K |
| Input modalities | text, image, file | audio, file, image, text, video |
| Tokenizer | Claude | Gemini |