Updated March 26, 2026· Based on independent benchmark data
Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) leads in intelligence with a score of 51.7 vs 44.6. GPT-5 Codex (high) is 2.4x cheaper at $1.25/1M tokens vs $3.00/1M. For speed, GPT-5 Codex (high) wins at 180 tok/s vs 71 tok/s.
| Metric | Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) | GPT-5 Codex (high) |
|---|---|---|
| Intelligence Score | 51.7 | 44.6 |
| Coding Score | 50.9 | 38.9 |
| Math Score | N/A | 98.7 |
| Speed (tok/s) | 71 tok/s | 180 tok/s |
| Latency (TTFT) | 32.16s | 9.17s |
| Input Price / 1M tokens | $3.00 | $1.25 |
| Output Price / 1M tokens | $15 | $10 |
| Context Window |
Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) outperforms GPT-5 Codex (high) on the intelligence index with a score of 51.7 compared to 44.6. For coding tasks, Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) has the edge with a coding score of 50.9 vs 38.9.
GPT-5 Codex (high) generates output significantly faster at 180 tok/s compared to Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)'s 71 tok/s, making it 2.5x faster for streaming responses. Time to first token is 9.17s for GPT-5 Codex (high) vs 32.16s for Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort), which affects perceived responsiveness in interactive applications.
GPT-5 Codex (high) is more affordable at $1.25/1M input tokens ($10/1M output), while Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) costs $3.00/1M input ($15/1M output). That makes Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) 2.4x more expensive per token, which can add up significantly at scale. For a typical workload of 100 requests per day at 2,000 tokens each, Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) would cost approximately $18.00/month vs $7.50/month for GPT-5 Codex (high) in input costs alone.
Choose Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) when you need higher intelligence (51.7), stronger coding performance (50.9). Choose GPT-5 Codex (high) when you need faster output (180 tok/s), lower cost.
Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) scores higher on coding benchmarks (50.9 vs 38.9), making it the better choice for programming tasks.
GPT-5 Codex (high) is cheaper at $1.25/1M input tokens vs $3.00/1M for Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort).
GPT-5 Codex (high) is faster, producing output at 180 tok/s compared to Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)'s 71 tok/s.
No, Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) 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. Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) scores higher on intelligence (51.7), but GPT-5 Codex (high) may be better for specific use cases like budget-conscious projects or speed-critical applications.