Back

OpenAI

GPT-5.4 mini

ReasoningClosed weights
Intelligence
19.7
Rank 252 of 744 measured models
Output speed
200 t/s
Rank 37 of 403 measured models
Price/1M
$1.69
Rank 448 of 659 measured models
Context window
400K
Rank 357 of 1199 measured models

GPT-5.4 mini (Medium) scores 19.7 on the Artificial Analysis Intelligence Index, ranking 252nd of 744 measured models. The closest models on this index are K-EXAONE 2.0 0803 (19.7), Step 3.7 Flash (19.5), and Claude 4.5 Sonnet (19.3). At $1.69 per million tokens (3:1 input/output blended price), it is cheaper than Claude 4.5 Sonnet ($6), more expensive than Step 3.7 Flash ($0.438). With an output speed of 200 t/s, it is faster than Step 3.7 Flash (185 t/s) and Claude 4.5 Sonnet (41 t/s).

Artificial Analysis Indices
Medium
Intelligence19.7
Rank 252 of 744 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
Standard Benchmarks
GPQA Diamond82.3%
Rank 182 of 619 measured models
HLE18.6%
Rank 233 of 664 measured models
IFBench64.8%
Rank 106 of 450 measured models
AA-LCR67.0%
Rank 245 of 575 measured models
TerminalBench Hard34.1%
Rank 91 of 432 measured models
τ²-Bench36.5%
Rank 243 of 440 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