GPT-4o-mini (2024-07-18)
GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs.
text + image + file → text · made by OpenAI
Sold by 4 ways
| Seller | Lane | Rate |
|---|---|---|
| Microsoft Azure AI cloud | standard | $0.17per Mtok in$0.66per Mtok out$0.083per Mtok cachedraw.githubusercontent.com · read 2026-09-16 |
| ElectronHub aggregator | standard | $0.1per Mtok in$0.4per Mtok outapi.electronhub.ai · read 2026-09-16 |
| Nous Research aggregator | standard | $0.15per Mtok in$0.6per Mtok out$0.075per Mtok cachedinference-api.nousresearch.com · read 2026-09-12 |
| OpenRouter aggregator | standard | $0.15per Mtok in$0.6per Mtok out$0.075per Mtok cachedopenrouter.ai · read 2026-08-24 |
Measured 29 standings
| Place | Board | Metric | Score |
|---|---|---|---|
| 1stof 113 | Piqa — Epoch AI | Score | 88.7 |
| 5thof 170 | Gsm8k — Epoch AI | EM | 0.913 |
| 7thof 136 | Bool q — Epoch AI | Score | 88.7 |
| 20thof 50 | Video mme — Epoch AI | Overall (no subtitles) | 0.648 |
| 24thof 217 | Mmlu — Epoch AI | EM | 0.818 |
| 26thof 36 | Balrog — Epoch AI | Average progress | 0.174 |
| 26thof 28 | LiveCodeBench Leaderboard (default window 8/1/2024–5/1/2025, 454 problems) | Pass@1 | 27.5 |
| 31stof 38 | Vpct — Epoch AI | Correct | 0.34 |
| 42ndof 49 | Lech mazur writing — Epoch AI | Mean score | 6.72 |
| 43rdof 54 | Live bench — Epoch AI | Global average | 41.26 |
| 59thof 108 | Math level 5 — Epoch AI | mean_score | 52.634 |
| 69thof 69 | Aider polyglot coding leaderboard | Percent correct | 3.6 |
| 72ndof 72 | Aider polyglot — Epoch AI | Percent correct | 3.6 |
| 79thof 80 | Simpleqa verified — Epoch AI | mean_score | 8.3 |
| 96thof 127 | Mystery game puzzles — Epoch AI | mean_score | 12 |
| 102ndof 102 | Simplebench — Epoch AI | Score (AVG@5) | 0.107 |
| 103rdof 106 | Frontiermath tiers 1 3 v2 — Epoch AI | mean_score | 0.702 |
| 108thof 123 | Lmca — Epoch AI | Score | 10.37 |
| 122ndof 152 | LMArena · Vision | Score (Elo) | 1098 |
| 130thof 161 | Dtbench — Epoch AI | Accuracy | 54.4 |
| 162ndof 170 | Weirdml — Epoch AI | Accuracy | 11.76 |
| 185thof 216 | ARC-AGI-2 (Public Eval) | Score (%) | 0 |
| 211thof 227 | Arc agi 2 — Epoch AI | Score | 0 |
| 217thof 232 | ARC-AGI-2 (Semi-Private) | Score (%) | 0 |
| 222ndof 222 | Chess puzzles — Epoch AI | mean_score | 0 |
| 247thof 290 | OTIS Mock AIME 2024-2025 | mean_score | 6.944 |
| 251stof 402 | LMArena · Text | Score (Elo) | 1318 |
| 272ndof 313 | GPQA Diamond | mean_score | 37.721 |
| 550thof 656 | Epoch capabilities index — Epoch AI | ECI Score | 126.56 |
About
GPT-4o-mini (2024-07-18) — a text model from OpenAI, sold by 4 companies from $0.1 in and $0.4 out per million tokens, placed 1st of 113 on Piqa — Epoch AI.
It takes text, images and files and returns text, with a context window of 128,000 tokens. It was published in July 2024, trained on material up to October 2023. Its sellers say it can call a tool. The catalogue files it under reasoning and code. Four companies sell it. The cheapest is $0.1 in and $0.4 out per million tokens at ElectronHub. It has been measured on 29 boards, and stands best at 1st of 113 on Piqa — Epoch AI.
Every current figure
- Maker
- OpenAI
- Register
- model
- Takes
- text + image + file
- Returns
- text
- Context
- 128,000 tokens
- Longest answer
- 16,384 tokens
- Published
- July 2024
- Knowledge to
- October 2023
- Licence
- not read
- Sellers
- 4
- Maker's own price
- not read
- Price
- $0.1 in and $0.4 out per million tokens — ElectronHub
- Boards
- 29
- Best place
- 1st of 113 — Piqa — Epoch AI
Known as 2 names
GPT-4O-mini-2024-07-18openai/gpt-4o-mini-2024-07-18