o1-mini
A small alternative to o1: a faster, more affordable o-series reasoning model designed to solve hard problems across domains; OpenAI recommends the newer o3-mini instead.
text → text · made by OpenAI
Sold by 2 ways
| Seller | Lane | Rate |
|---|---|---|
| Microsoft Azure AI cloud | batch | $0.6per Mtok in$2.42per Mtok outraw.githubusercontent.com · read 2026-09-16 |
| Microsoft Azure AI cloud | standard | $1.21per Mtok in$4.84per Mtok out$0.6per Mtok cachedraw.githubusercontent.com · read 2026-09-16 |
Measured 19 standings
| Place | Board | Metric | Score |
|---|---|---|---|
| 17thof 22 | Cybench — Epoch AI | Unguided % Solved (medium) | 0.1 |
| 21stof 54 | Live bench — Epoch AI | Global average (medium) | 57.76 |
| 24thof 108 | Math level 5 — Epoch AI | mean_score (high) | 89.181 |
| 43rdof 49 | Lech mazur writing — Epoch AI | Mean score (medium) | 6.49 |
| 48thof 69 | Aider polyglot coding leaderboard | Percent correct | 32.9 |
| 52ndof 72 | Aider polyglot — Epoch AI | Percent correct | 32.9 |
| 88thof 101 | FrontierMath (Tiers 1-3) | mean_score (medium) | 1.724 |
| 99thof 102 | Simplebench — Epoch AI | Score (AVG@5) (medium) | 0.181 |
| 135thof 170 | Weirdml — Epoch AI | Accuracy (medium) | 36.32 |
| 186thof 290 | OTIS Mock AIME 2024-2025 | mean_score (high) | 46.944 |
| 190thof 205 | ARC-AGI-1 (Public Eval) | Score (%) | 13.16 |
| 191stof 216 | ARC-AGI-2 (Public Eval) | Score (%) | 0 |
| 200thof 227 | Arc agi 2 — Epoch AI | Score | 0.83 |
| 200thof 313 | GPQA Diamond | mean_score (high) | 62.374 |
| 202ndof 232 | ARC-AGI-2 (Semi-Private) | Score (%) | 0.83 |
| 204thof 230 | ARC-AGI-1 (Semi-Private) | Score (%) | 14 |
| 217thof 244 | Arc agi — Epoch AI | Score (medium) | 14 |
| 223rdof 402 | LMArena · Text | Score (Elo) | 1337 |
| 492ndof 656 | Epoch capabilities index — Epoch AI | ECI Score (high) | 135.84 |
About
o1-mini — a text model from OpenAI, sold by one company from $1.21 in and $4.84 out per million tokens, placed 24th of 108 on Math level 5 — Epoch AI.
It takes text and returns text, with a context window of 128,000 tokens. It was published in 2025, trained on material up to January 2025. The catalogue files it under reasoning and code. Only Microsoft Azure AI sells it, at $1.21 in and $4.84 out per million tokens. Beside the standard rate there is a batch lane. It has been measured on 19 boards, and stands best at 24th of 108 on Math level 5 — Epoch AI.
Every current figure
- Maker
- OpenAI
- Register
- model
- Takes
- text
- Returns
- text
- Context
- 128,000 tokens
- Longest answer
- 65,536 tokens
- Published
- 2025
- Knowledge to
- January 2025
- Licence
- not read
- Sellers
- 1
- Maker's own price
- not read
- Price
- $1.21 in and $4.84 out per million tokens — Microsoft Azure AI
- Boards
- 19
- Best place
- 24th of 108 — Math level 5 — Epoch AI
Known as 1 name
o1 mini