benchr
Compare · Side-by-side
Claude Sonnet 4.6 vs GPT-5
Detailed comparisons of token costs, context window limits, benchmark performance, and speed.
| Dimension | Claude Sonnet 4.6 | GPT-5 |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Input Tokens Cost / 1M | $3.000 | $1.250 |
| Output Tokens Cost / 1M | $15.000 | $10.000 |
| Cached Input / 1M | $0.300 | — |
| Max Context Window | 1,000,000 tokens | 400,000 tokens |
| SWE-bench Verified | 79.6% | 74.9% |
| GPQA Diamond | 89.9% | — |
Summary Recommendations
Choose Claude Sonnet 4.6 if you need: Production default.
Choose GPT-5 if you need: Production workhorse at a rational price.