o3-mini
o3-mini is OpenAI's most recent small reasoning model, providing high intelligence at the same cost and latency targets of o1-mini.
text → text · made by OpenAI
$1.1→$4.4per Mtok in / out
OpenAI · 8 sellers
Sold by 14 ways
| Seller | Lane | Rate |
|---|---|---|
| Microsoft Azure AI cloud | batch | $0.6per Mtok in$2.42per Mtok outraw.githubusercontent.com · read 2026-09-16 |
| DigitalOcean cloud | standard | $1.1per Mtok in$4.4per Mtok out$0.55per Mtok cachedmodels.dev · read 2026-09-16 |
| Microsoft Azure AI cloud | standard | $1.1per Mtok in$4.4per Mtok out$0.55per Mtok cachedmodels.dev · read 2026-09-16 |
| OpenAI api | standard | $1.1per Mtok in$4.4per Mtok out$0.55per Mtok cacheddevelopers.openai.com · read 2026-08-24 |
| Nous Research 2 ways | $1.1per Mtok in$4.4per Mtok out$0.55per Mtok cached | |
| Nous Research aggregator | batch | $0.55per Mtok in$2.2per Mtok out$0.28per Mtok cachedinference-api.nousresearch.com · read 2026-09-12 |
| Nous Research aggregator | standard | $1.1per Mtok in$4.4per Mtok out$0.55per Mtok cachedinference-api.nousresearch.com · read 2026-09-12 |
| OpenRouter 2 ways | $1.1per Mtok in$4.4per Mtok out$0.55per Mtok cached | |
| OpenRouter aggregator | batch | $0.55per Mtok in$2.2per Mtok out$0.28per Mtok cachedopenrouter.ai · read 2026-08-24 |
| OpenRouter aggregator | standard | $1.1per Mtok in$4.4per Mtok out$0.55per Mtok cachedopenrouter.ai · read 2026-08-24 |
| Requesty 2 ways | $0.99per Mtok in$3.96per Mtok out$0.49per Mtok cached | |
| Requesty aggregatornot on the seller's list today | openai low | $0.99per Mtok in$3.96per Mtok out$0.49per Mtok cachedrouter.requesty.ai · read 2026-08-25 |
| Requesty aggregatornot on the seller's list today | openai medium | $0.99per Mtok in$3.96per Mtok out$0.49per Mtok cachedrouter.requesty.ai · read 2026-08-25 |
| Requesty aggregator | openai | $1.1per Mtok in$4.4per Mtok out$0.55per Mtok cachedrouter.requesty.ai · read 2026-08-28 |
| Requesty aggregator | openai-responses | $1.1per Mtok in$4.4per Mtok out$0.55per Mtok cachedrouter.requesty.ai · read 2026-08-28 |
| ElectronHub aggregator | standard | $1.1per Mtok in$4.4per Mtok outapi.electronhub.ai · read 2026-09-16 |
| Vercel AI Gateway aggregator | standard | $1.1per Mtok in$4.4per Mtok out$0.55per Mtok cachedai-gateway.vercel.sh · read 2026-08-25 |
Measured 37 standings
| Place | Board | Metric | Score |
|---|---|---|---|
| 4thof 54 | Live bench — Epoch AI | Global average (high) | 75.88 |
| 6thof 15 | Cad eval — Epoch AI | Overall pass (%) (medium) | 0.54 |
| 9thof 28 | LiveCodeBench Leaderboard (default window 8/1/2024–5/1/2025, 454 problems) | Pass@1 (high) | 67.4 |
| 10thof 108 | Math level 5 — Epoch AI | mean_score (high) | 96.488 |
| 11thof 22 | Cybench — Epoch AI | Unguided % Solved (medium) | 0.225 |
| 15thof 28 | LiveCodeBench Leaderboard (default window 8/1/2024–5/1/2025, 454 problems) | Pass@1 (med) | 63 |
| 17thof 25 | AgentBench Leaderboard (new) | Success Rate (pass@1) AVG | 40.9 |
| 18thof 28 | LiveCodeBench Leaderboard (default window 8/1/2024–5/1/2025, 454 problems) | Pass@1 (low) | 57 |
| 27thof 72 | Aider polyglot — Epoch AI | Percent correct (high) | 60.4 |
| 30thof 42 | Fictionlivebench — Epoch AI | 120k token score (medium) | 0.438 |
| 32ndof 69 | Aider polyglot coding leaderboard | Percent correct (medium) | 53.8 |
| 36thof 38 | Gso — Epoch AI | Score OPT@1 (low) | 0.013 |
| 37thof 81 | Forecastbench — Epoch AI | Overall score | 59.6 |
| 45thof 72 | Frontiermath tier 4 — Epoch AI | mean_score (high) | 4.167 |
| 47thof 49 | Lech mazur writing — Epoch AI | Mean score (high) | 6.17 |
| 52ndof 101 | FrontierMath (Tiers 1-3) | mean_score (high) | 12.414 |
| 62ndof 62 | Frontiermath tier 4 v2 — Epoch AI | mean_score (high) | 0 |
| 71stof 80 | Simpleqa verified — Epoch AI | mean_score (high) | 15.3 |
| 82ndof 106 | Frontiermath tiers 1 3 v2 — Epoch AI | mean_score (high) | 18.596 |
| 88thof 161 | Dtbench — Epoch AI | Accuracy (high) | 68.8 |
| 90thof 123 | Lmca — Epoch AI | Score (high) | 18.97 |
| 92ndof 102 | Simplebench — Epoch AI | Score (AVG@5) (high) | 0.228 |
| 94thof 170 | Weirdml — Epoch AI | Accuracy (high) | 43.7 |
| 103rdof 222 | Chess puzzles — Epoch AI | mean_score (high) | 17 |
| 111thof 127 | Mystery game puzzles — Epoch AI | mean_score (high) | 7 |
| 116thof 290 | OTIS Mock AIME 2024-2025 | mean_score (high) | 76.944 |
| 119thof 168 | Scicode — Epoch AI | Score (high) | 39.786 |
| 131stof 177 | Critpt — Epoch AI | Accuracy (high) | 0.286 |
| 138thof 313 | GPQA Diamond | mean_score (high) | 77.02 |
| 146thof 205 | ARC-AGI-1 (Public Eval) | Score (%) (high) | 46.58 |
| 157thof 227 | Arc agi 2 — Epoch AI | Score (high) | 2.99 |
| 158thof 232 | ARC-AGI-2 (Semi-Private) | Score (%) (high) | 2.99 |
| 160thof 230 | ARC-AGI-1 (Semi-Private) | Score (%) (high) | 34.5 |
| 170thof 244 | Arc agi — Epoch AI | Score (high) | 34.5 |
| 192ndof 216 | ARC-AGI-2 (Public Eval) | Score (%) (low) | 0 |
| 207thof 402 | LMArena · Text | Score (Elo) | 1348 |
| 426thof 656 | Epoch capabilities index — Epoch AI | ECI Score (high) | 140.38 |
About
o3-mini — a text model from OpenAI, sold by 8 companies from $1.1 in and $4.4 out per million tokens, placed 4th of 54 on Live bench — Epoch AI.
It takes text and returns text, with a context window of 200,000 tokens. It was published in December 2024, trained on material up to May 2024. Its sellers say it can reason step by step and call a tool. The catalogue files it under reasoning, code and agents. Eight companies sell it. The cheapest is $1.1 in and $4.4 out per million tokens at DigitalOcean. Beside the standard rate there are batch and separately routed lanes. It has been measured on 35 boards, and stands best at 4th of 54 on Live bench — Epoch AI.
Every current figure
- Maker
- OpenAI
- Register
- model
- Takes
- text
- Returns
- text
- Context
- 200,000 tokens
- Longest answer
- 131,072 tokens
- Published
- December 2024
- Knowledge to
- May 2024
- Licence
- not read
- Sellers
- 8
- Price
- $1.1 in and $4.4 out per million tokens — DigitalOcean
- Boards
- 35
- Best place
- 4th of 54 — Live bench — Epoch AI
Known as 11 names
o3-mini (2025-01-31)O3-Mini-2025-01-31 (High)O3-Mini-2025-01-31 (Low)O3-Mini-2025-01-31 (Med)o3 mini 0131openai/o3-miniopenai/o3-mini:batchopenai-responses/o3-miniopenai/o3-mini:lowopenai/o3-mini:mediumOpenAI: o3 Mini (batch)