Index · comparison
Claude Sonnet 5 (Anthropic) vs GPT-5.6 Sol (OpenAI)
GPT-5.6 Sol (OpenAI) is 50% cheaper than Claude Sonnet 5 (Anthropic) ($6.00 vs $12.00 blended per 1M tokens).
Claude Sonnet 5 (Anthropic) on Anthropic vs GPT-5.6 Sol (OpenAI) on OpenAI — mid tier workhorses. Every figure below is the value shown on each model's own endpoint page, with its source and verification date.
| Metric | Claude Sonnet 5 (Anthropic) | GPT-5.6 Sol (OpenAI) |
|---|---|---|
| Input $/1M | $2 | $1 |
| Output $/1M | $10 | $5 |
| Blended $/1M | $12 | $6 |
| Context window | 200K | 400K |
| Max output | 64K | 128K |
| Intelligence index | 55.3 | 58.9 |
Highlighted cell = better on that row (lower price/latency, higher context/intelligence/throughput). "—" = not measured for that model.
Which to pick
On blended token price, GPT-5.6 Sol (OpenAI) is the cheaper of the two ($6 vs $12 per 1M). Price alone rarely settles it: check the intelligence index and latency rows above against your workload, and open each endpoint page for the full price history and provider terms.
Sources: Claude Sonnet 5 (Anthropic) endpoint · GPT-5.6 Sol (OpenAI) endpoint · Claude Sonnet 5 (Anthropic) price verified 2026-09-05 · GPT-5.6 Sol (OpenAI) price verified 2026-09-05