Side-by-side comparison of AI model specs, benchmarks, and pricing.
| Spec | GPT-5.3 Codex OpenAI | o4-mini OpenAI |
|---|---|---|
| Tier | frontier | frontier |
| Release Date | 2026-02-05 | 2025-04-16 |
| Context Window | 400K | 200K |
| Max Output | 128K | 100K |
| Input Price (/1M) | $1.75 | $1.1 |
| Output Price (/1M) | $14 | $4.4 |
| Arena Elo | 1,480 | 1,380 |
| MMLU | 93% | 86.8% |
| GPQA | 81% | 73.4% |
| MATH | 96% | 93.4% |
| HumanEval | 93% | 93.6% |
| SWE-bench | 80% | 68.5% |
| AIME | 94% | 93.4% |
| SimpleQA | 58% | 40.3% |
| Capabilities | vision, tool-use, code, reasoning, agentic | vision, tool-use, code, reasoning |