The Model Index
Models/OpenAI

GPT-5.4 Mini

OpenAIClosed weights

GPT-5.4 Mini is a closed AI model from OpenAI with an index score of 147 (GPT-4 = 100), rank 77 of 217.

147131–163Index score#77 of 217 models
$0.75/ $4.5Price per million tokensinput / output · via OpenRouter
400KContext windowtokens it reads at once
17 Mar 2026Released8 months ago

Where it sitson the frontier

801001201401601802002023202420252026GPT-4 at launch = 100GPT-5.4 Mini

o3 reached this score 11 months earlier.

Test resultsby area · tick = best by any model

Reasoning & knowledgeevidence
GPQA Diamond (Epoch AI)
87%
Humanity's Last Exam
no result
Mathsevidence
FrontierMath (tiers 1-3)
51%
Codingevidence
SWE-bench Verified
no result
Terminal-Bench 4.0
3%
Agentsevidence
APEX-Agents
no result
Professional workevidence
Finance Agent (v2)
45%
Novel problemsevidence
ARC-AGI-2
19%
Long tasksno evidence yet
METR time horizon
no result

Third-party results only, at the model’s best standard setting. Hover a bar for who ran it.

What others sayrank on 10 leaderboards

Best attop three among current models

Not in the top three among current models on any category we track.

Where to use itprices checked 18 Nov 2026

Host prices from OpenRouter; going direct can differ.

Facts and sources

Released17 Mar 2026
Announcementnot linked yet
Release entry–
Scored asGPT-5.4 Mini
Output speed–
Hugging Faceclosed weights

OpenAI releasesaround this one

This page as a chartDownload PNG