Side-by-side comparison of AI model specs, benchmarks, and pricing.
| Spec | Gemini 3.1 Pro Google | Grok 4.1 xAI |
|---|---|---|
| Tier | frontier | frontier |
| Release Date | 2026-02-19 | 2025-11-17 |
| Context Window | 1049K | 2000K |
| Max Output | 66K | 16K |
| Input Price (/1M) | $2 | $0.2 |
| Output Price (/1M) | $12 | $0.5 |
| Arena Elo | 1,500 | 1,465 |
| MMLU | 92.6% | 90.2% |
| GPQA | 94.3% | 88% |
| MATH | 96.8% | — |
| HumanEval | 94.6% | 92% |
| SWE-bench | 80.6% | 65% |
| AIME | 91.2% | 78% |
| SimpleQA | 79.6% | 38% |
| Capabilities | vision, tool-use, code, reasoning, agentic, audio | vision, tool-use, code, reasoning |