Updated March 26, 2026· Based on independent benchmark data
GPT-5.3 Codex (xhigh) leads in intelligence with a score of 54.0 vs 51.7. GPT-5.3 Codex (xhigh) is 1.7x cheaper at $1.75/1M tokens vs $3.00/1M.
| Metric | GPT-5.3 Codex (xhigh) | Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) |
|---|---|---|
| Intelligence Score | 54.0 | 51.7 |
| Coding Score | 53.1 | 50.9 |
| Math Score | N/A | N/A |
| Speed (tok/s) | 72 tok/s | 71 tok/s |
| Latency (TTFT) | 83.91s | 32.16s |
| Input Price / 1M tokens | $1.75 | $3.00 |
| Output Price / 1M tokens | $14 | $15 |
| Context Window |
GPT-5.3 Codex (xhigh) outperforms Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) on the intelligence index with a score of 54.0 compared to 51.7. For coding tasks, GPT-5.3 Codex (xhigh) has the edge with a coding score of 53.1 vs 50.9.
Both models deliver similar output speeds: GPT-5.3 Codex (xhigh) at 72 tok/s and Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) at 71 tok/s. Time to first token is 32.16s for Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) vs 83.91s for GPT-5.3 Codex (xhigh), which affects perceived responsiveness in interactive applications.
GPT-5.3 Codex (xhigh) is more affordable at $1.75/1M input tokens ($14/1M output), while Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) costs $3.00/1M input ($15/1M output). For a typical workload of 100 requests per day at 2,000 tokens each, GPT-5.3 Codex (xhigh) would cost approximately $10.50/month vs $18.00/month for Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) in input costs alone.
Choose GPT-5.3 Codex (xhigh) when you need higher intelligence (54.0), stronger coding performance (53.1), lower cost.
GPT-5.3 Codex (xhigh) scores higher on coding benchmarks (53.1 vs 50.9), making it the better choice for programming tasks.
GPT-5.3 Codex (xhigh) is cheaper at $1.75/1M input tokens vs $3.00/1M for Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort).
GPT-5.3 Codex (xhigh) is faster, producing output at 72 tok/s compared to Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)'s 71 tok/s.
No, GPT-5.3 Codex (xhigh) does not support image input. Neither model supports image input.
Data last synced: March 26, 2026
| N/A |
| N/A |
| Max Output Tokens | N/A | N/A |
| Input Modalities | Text | Text |
| Output Modalities | Text | Text |
| Free Tier | No | No |
It depends on your priorities. GPT-5.3 Codex (xhigh) scores higher on intelligence (54.0), but Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) may be better for specific use cases like budget-conscious projects or speed-critical applications.