Back

OpenAI

o3-mini

ReasoningClosed weights
Intelligence
11.0
Rank 406 of 744 measured models
Coding
16.3
Rank 240 of 298 measured models
Output speed
187 t/s
Rank 41 of 403 measured models
Price/1M
$1.93
Rank 459 of 659 measured models
Context window
200K
Rank 618 of 1199 measured models

o3-mini (High) scores 11.0 on the Artificial Analysis Intelligence Index, ranking 406th of 744 measured models. The closest models on this index are GLM-4.5-Air (11.1), Granite 4.2 8B (11.1), and Mi:dm K 2.5 Pro (11.0). At $1.93 per million tokens (3:1 input/output blended price), it is more expensive than GLM-4.5-Air ($0.372) and Granite 4.2 8B ($0.107). With an output speed of 187 t/s, it is faster than GLM-4.5-Air (71.1 t/s) and Granite 4.2 8B (114 t/s).

Artificial Analysis Indices
High
Intelligence11.0
Rank 406 of 744 measured models
Coding16.3
Rank 240 of 298 measured models
Agentic0.9
Rank 166 of 189 measured models
Reasoning levels
Each point is a reasoning level of this model: more reasoning often raises the score but costs more. Select a point to open that level.
Intelligence Indexmost efficient ↗Price / 1M tokens
Capability indexes
Performance in specific industries, measured by Artificial Analysis
Finance & Accounting10.0
Rank 187 of 218 measured models
Strategy & Ops6.4
Rank 190 of 218 measured models
Legal11.2
Rank 191 of 218 measured models
Engineering13.4
Rank 190 of 217 measured models
Economics16.0
Rank 197 of 225 measured models
Standard Benchmarks
MMLU Pro80.2%
Rank 113 of 349 measured models
GPQA Diamond77.3%
Rank 237 of 619 measured models
HLE12.0%
Rank 293 of 664 measured models
LiveCodeBench73.4%
Rank 55 of 343 measured models
SciCode42.8%
Rank 157 of 231 measured models
MATH-50098.5%
Rank 10 of 202 measured models
AIME 202486.0%
Rank 13 of 194 measured models
IFBench67.1%
Rank 90 of 450 measured models
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.

Preferences and analytics

Rybbit: visits, performance, clicks, copies, forms and session replay (masked inputs), with your permission. Refusing does not limit the site.

Privacy

Allow navigation analytics?

With your permission, Rybbit measures visits, performance and interactions (clicks, copies, forms), including session replay, to improve BenchSift. Input values are masked in replays. Refusing does not limit the site.

Your choice is stored for 180 days. Change it at any time using “Privacy preferences” in the footer. Withdrawal reloads the page.

No choice stored: Rybbit is disabled.

Privacy