Head to head
o3 vs Claude Sonnet 4
o3: legacyClaude Sonnet 4: retiredheavy tier
Verdict · modelled basis
o3 costs 2.2x less per solved task.
On output token price alone, o3 looks 1.9x cheaper (1.5x on input). The ranking holds, but the margin changes.
Source: Aider polyglot benchmark · 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
modelled basis · lower is better
| o3 | Claude Sonnet 4 | |
|---|---|---|
| Input $/Mtok | $2 | $3 |
| Output $/Mtok | $8 | $15 |
| Cached input $/Mtok | $0.5 | $0.3 |
| Pass rate | 81% | 61% |
| Pass rate source | Aider | Aider |
| Cost basis | modelled | modelled |
| Cost per solved task | $1.99 | $4.40 |
verified 2026-08-21 · heavy task tier; measured rows ignore the tier· Methodology· o3· Claude Sonnet 4