anthropic--claude-3-sonnetThe provider has deprecated this model. It still responds, but it is not a good choice for new projects.
| Provider | Input, $ per 1M | Output, $ per 1M | In our data since |
|---|---|---|---|
| Cloudflare AI Gateway | $3 | $15 | 1 Aug 2026 |
| Google Vertex AI | $3 | $15 | 1 Aug 2026 |
| SAP AI Core | $3 | $15 | 1 Aug 2026 |
Only the base tier and only per-token prices are shown. Batch, discounted and cached rates, as well as prices per image or per second of video, do not go into this table: they cannot stand in the same column as a price per million tokens.
The date in the last column is the day this price first entered our collection. The price may well be older: before that day we simply were not recording it. It has not changed since — otherwise a new row with a new date would stand in its place.
| Task set | Result | Run conditions | Measured by |
|---|---|---|---|
| Arena Score in French | 1,227.84 1,207–1,249 | — | — |
| Arena Score, multi-turn dialogue | 1,226.91 1,219–1,235 | — | — |
| Arena Score in English | 1,226.25 1,221–1,231 | — | — |
| Arena Score in Russian | 1,225.66 1,216–1,235 | — | — |
| Arena Score, programming | 1,223.21 1,216–1,231 | — | — |
| Arena Score, overall | 1,218.2 1,214–1,222 | — | — |
| Arena Score, mathematics | 1,213.44 1,205–1,221 | — | — |
| Arena Score, long queries | 1,211.28 1,203–1,220 | — | — |
| Arena Score in Spanish | 1,203.52 1,182–1,225 | — | — |
| Arena Score, instruction following | 1,199.84 1,194–1,205 | — | — |
| Arena Score, hard prompts | 1,197.3 1,191–1,203 | — | — |
| Arena Score, creative writing | 1,186.58 1,178–1,195 | — | — |
| Arena Score, expert questions | 1,172.02 1,160–1,184 | — | — |
| Arena Score, working with images | 984.05 973–995 | — | — |
| GPQA Diamond — graduate-level questions | 20.79 % | · with a tuned harness | Epoch evaluations |
| MATH, difficulty level five | 18.17 % | · with a tuned harness | Epoch evaluations |
| WeirdML — unusual machine learning tasks | 10.16 % | — | WeirdML Leaderboard |
| Mock AIME 2024–2025 — olympiad problems | 2.4 % | · with a tuned harness | Epoch evaluations |