o3
Model ID
- Context
- 200K
- 100K max output
- Input price
- $2.00
- per 1M tokens · OpenAI API
- Output price
- $8.00
- per 1M tokens
- Epoch Capabilities Index
- 146.9
- #55 of 105 listed
- LMArena Text
- 1,432
- #75 of 149 listed
Where to run it
3 offerings · prices as listed| Meter | Tier | Price |
|---|---|---|
| Input | Standard | $2.00 / 1M tokens |
| Batch | $1.00 / 1M tokens | |
| Flex tier | $1.00 / 1M tokens | |
| Fast tier | $3.50 / 1M tokens | |
| Output | Standard | $8.00 / 1M tokens |
| Batch | $4.00 / 1M tokens | |
| Flex tier | $4.00 / 1M tokens | |
| Fast tier | $14.00 / 1M tokens | |
| Cache read | Standard | $0.50 / 1M tokens |
| Flex tier | $0.25 / 1M tokens | |
| Fast tier | $0.88 / 1M tokens | |
Azure OpenAI in Foundry (opens in a new tab) o3-2025-04-16 | ||
| Input | Standard | $2.00 / 1M tokens |
| Batch | $1.00 / 1M tokens | |
| Output | Standard | $8.00 / 1M tokens |
| Batch | $4.00 / 1M tokens | |
| Cache read | Standard | $0.50 / 1M tokens |
Azure OpenAI in Foundry (opens in a new tab) o3-2025-04-16 | ||
| Input | Standard | $2.20 / 1M tokens |
| Batch | $1.10 / 1M tokens | |
| Output | Standard | $8.80 / 1M tokens |
| Batch | $4.40 / 1M tokens | |
| Cache read | Standard | $0.55 / 1M tokens |
Changes
All changes →- Available → Deprecated
- Now priced on Azure OpenAI in Foundry, Data zone Input $2.20 · Cache read $0.55 · Output $8.80 · Input (Batch) $1.10 · +1 more
- Now priced on Azure OpenAI in Foundry Input $2.00 · Cache read $0.50 · Output $8.00 · Input (Batch) $1.00 · +1 more
Cited scores
From third-party leaderboards, not measured by us- Epoch Capabilities Index 146.9
#55 of 105 listed models · CI 144.9–148.6
- LMArena Text 1,432
#75 of 149 listed models · CI 1,428–1,435 · 58,583 votes
- Epoch AI · FrontierMath Tiers 1–3 33.3%
#55 of 74 listed models · high effort · ±2.8
Also listed: medium effort 29.8% · low effort 19.3%
- Epoch AI · GPQA Diamond 81.8%
#64 of 128 listed models · high effort · ±2.13
Also listed: medium effort 80.8% · low effort 79.8%
- Epoch AI · SWE-bench Verified 62.3%
#25 of 30 listed models · medium effort · ±2.21
- LMArena Text · Chinese 1,464
#77 of 149 listed models · CI 1,453–1,474 · 3,848 votes
- LMArena Text · Coding 1,460
#87 of 149 listed models · CI 1,454–1,466 · 11,534 votes
- LMArena Vision 1,214
#56 of 87 listed models · CI 1,208–1,221 · 45,593 votes
- Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Oct 3, 2026
- LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Oct 2, 2026
- Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Oct 1, 2026
- changes we made