| 100,000 tokens |
| Knowledge cutoff | Oct 1, 2023 |
|---|
| Input | Text |
|---|
| Reasoning effort | low, medium, high |
|---|
| API price per 1M tokens (input / cached / output) | $1.10 / $0.55 / $4.40 |
|---|
From OpenAI's API model page (the model is no longer current); the page publishes no benchmark figures, so no chart is saved. · source
Estimate from the version's published specs, benchmarks and reviews, 0-5; scores are only comparable within one criteria year. Scored as launched on OpenAI's API model page (opened this run): 200,000-token context = 2; list price $1.10 input / $4.40 output per 1M = $1.93 blended at 3:1 = 4.5; text-only input = 1. Reasoning and coding are null: the Artificial Analysis index at launch and an independent Terminal-Bench figure were not opened this run, and OpenAI's own announcement returned no readable page. Speed is null (OpenAI's 7.7 s vs 10.16 s response time against o1-mini is not tokens per second). Three independent pages were found: Tom's Guide hands-on test, Epoch AI's independent benchmark page, and The Decoder's launch report (which largely relays OpenAI's figures). developers.openai.com
The o3 model nailed the answer extremely easily, using both high and low reasoning.
At the time of its release, it ranked 2nd of 91 tracked models.
Writing, text and creative tasks don't benefit from this training method, which is why OpenAI recommends the o-series primarily for logic tasks and analyses.
| OpenAI launches new reasoning model o3-mini for free ChatGPT and API · Jan 31, 2025 | 0.20 |
|---|---|
| I put OpenAI's new o3-mini model to the test — and I'm shocked by the results | Tom's Guide · Feb 6, 2025 | 0.80 |
Photos from readers, kept apart from the pin's sourced media. Rate what you ate or saw, and on a restaurant pin tag the dishes right on your photo.
The first entry is always the pin's source. Overall confidence is a weighted average of how firmly each reference supports the start and end times used above; a reference counts half as much for every 180 days older than the newest.
OpenAI's own model release notes date this entry January 31, 2025
The OpenAI[1] Daily Brief's episode "o3-Mini Model, U.S. Energy Collaboration, and SoftBank's Stargate Investment" (2025-01-31) says: "AI reasoning faster and more accessible than ever!? Welcome to The OpenAI Weekly Brief, your go-to for the latest AI updates. Today is Friday, January 31st , 2025. Here’s what you need to know about OpenAI's groundbreaking launch of the o3-mini reasoning model. Let’s dive in. OpenAI has just unveiled its"
Archived copy of OpenAI's[1] o3-mini announcement, which says it is "our first small reasoning model that supports ... function calling, Structured Outputs, and developer messages", offers low, medium and high reasoning effort, and averaged "7.7 seconds compared to 10.16 seconds" for o1-mini.
Something missing or wrong? Say it in your own words: a link that backs this pin up, a different start or end date and why, or a fact it lacks or gets wrong. The AI checks it against this pin's sources, searches for better ones, and adds any page that backs you up. The pin's own sources still count most. A picture that shows something else, or shows it badly, is looked at too, and moved down or replaced.