Claude Sonnet 5 vs GPT-5.1-Codex
Side-by-side API pricing and Artificial Analysis performance for Claude Sonnet 5 (Anthropic) and GPT-5.1-Codex (OpenAI). Green cells mark the better value in each row.
Prices updated Sep 15, 2026, 7:29 PM UTC · refreshed hourly
| Claude Sonnet 5 Anthropic | GPT-5.1-Codex OpenAI | |
|---|---|---|
| Pricing | ||
| Input $/1M | $2.00 | $1.25 |
| Output $/1M | $10.00 | $10.00 |
| Blended $/1M (3:1) | $4.00 | $3.44 |
| RAG example (30K in / 2K out) | $0.08 | $0.0575 |
| Quality & speed | ||
| Intelligence Index | 38.4 | 23.7 |
| Coding Index | 71.5 | — |
| Agentic Index | 44.3 | — |
| Output speed (tok/s) | 84 | 0 |
| Time to first token | 125.08s | 0s |
| Specs | ||
| Context window | 1M | 400K |
| Input modalities | text, image, file | text, image |
FAQ
Which is cheaper, Claude Sonnet 5 or GPT-5.1-Codex?
GPT-5.1-Codex has the lower blended API price at $3.44 per 1M tokens (3:1 input:output mix), versus $4.00 for the other model. Prices are live from OpenRouter and refresh about hourly.
Which is better for RAG workloads?
For a typical RAG request (30K input / 2K output tokens), GPT-5.1-Codex costs about $0.0575 per request versus $0.08. Also compare context windows: Claude Sonnet 5 offers 1M and GPT-5.1-Codex offers 400K.
Which scores higher on Artificial Analysis benchmarks?
Claude Sonnet 5 leads on the Artificial Analysis Intelligence Index (38.4 vs 23.7). Check coding and agentic indices on this page for workload-specific tradeoffs.