Head to head
GPT-5.3 Codex vs Claude Sonnet 4
GPT-5.3 Codex: currentClaude Sonnet 4: retiredheavy tier
No verdict
Cost per solved task needs a published pass rate for both models. At least one has none, so it is reported as missing rather than estimated.
verified 2026-08-21 · prices from each provider's pricing page
Input $/M tokens
lower is better
Output $/M tokens
lower is better
Cost per solved task
needs a pass rate for both
| GPT-5.3 Codex | Claude Sonnet 4 | |
|---|---|---|
| Input $/Mtok | $1.75 | $3 |
| Output $/Mtok | $14 | $15 |
| Cached input $/Mtok | $0.175 | $0.3 |
| Pass rate | missing | 61% |
| Pass rate source | missing | Aider |
| Cost basis | missing | modelled |
| Cost per solved task | missing | $4.40 |
verified 2026-08-21 · heavy task tier; measured rows ignore the tier· Methodology· GPT-5.3 Codex· Claude Sonnet 4