GPT-5.4 Mini

ModelActive

by OpenAI · family “gpt-mini”

GPT-5.4 mini brings the core capabilities of GPT-5.4 to a faster, more efficient model optimized for high-throughput workloads. It supports text and image inputs with strong performance across reasoning, coding,... (description from the OpenRouter listing)

in: filein: imagein: textout: textReasoningTool useStructured output

Epoch Capabilities Index

148.8 #56 of 268

90% CI 146.9 – 150.4

Listing ↗

Released

17 Mar 2026

Context window

400K tokens

Max output

128K tokens

Input $ / 1M tokens

$0.75

Output $ / 1M tokens

$4.50

Cached input $ / 1M

$0.075

Each benchmark shown separately with its own source

Benchmark results (9)

All benchmarks

Benchmark	Domain	Score	vs best recorded	Setting	Run	Source
Chess Puzzles	games	18.0% ±3.9	25%	high	15 Apr 2026	Eval log ↗
SimpleQA Verified	knowledge	29.4% ±1.4	39%	high	27 Aug 2026	Eval log ↗
FrontierMath-Tier-4-2025-07-01-Privatesuperseded	math	2.1% ±2.1	4%	high	15 Apr 2026	Epoch ↗
ARC-AGI-2	reasoning	13.2%	14%	high	—	External ↗
FrontierMath-2025-02-28-Privatesuperseded	math	28.3% ±2.1	54%	high	15 Apr 2026	Epoch ↗
WeirdML	coding	60.3% ±0.0	64%	high	—	External ↗
OTIS Mock AIME 2024-2025	math	87.2% ±4.3	87%	high	15 Apr 2026	Eval log ↗
GPQA diamond	science	83.6% ±2.2	87%	high	15 Apr 2026	Eval log ↗
ARC-AGI	reasoning	58.0%	59%	high	—	External ↗

Source: Epoch AI Benchmarking Hub (CC BY 4.0). “External” rows are leaderboard results Epoch collects from third parties. Best reported setting per benchmark is shown.

API list price over time

Price history

$ per 1M tokens

7 observations (OpenRouter listing + Internet Archive snapshots). History accumulates with every ingest run; a single point means no change has been observed yet.