Side-by-side comparison of AI model specs, benchmarks, and pricing.
| Spec | Claude Sonnet 4.5 Anthropic | Grok 4 xAI |
|---|---|---|
| Tier | frontier | frontier |
| Release Date | 2025-09-29 | 2025-07-09 |
| Context Window | 200K | 256K |
| Max Output | 64K | 8K |
| Input Price (/1M) | $3 | $3 |
| Output Price (/1M) | $15 | $15 |
| Arena Elo | 1,450 | 1,410 |
| MMLU | 88.7% | 87.5% |
| GPQA | 83.4% | 87.5% |
| MATH | 87% | 91.7% |
| HumanEval | 95% | 90% |
| SWE-bench | 55.8% | 48% |
| AIME | 58% | 72% |
| SimpleQA | 30.8% | 34.2% |
| Capabilities | vision, tool-use, code, extended-thinking | vision, tool-use, code, reasoning |