o4 Mini

ModelActive

by OpenAI · family “o4-mini”

OpenAI o4-mini is a compact reasoning model in the o-series, optimized for fast, cost-efficient performance while retaining strong multimodal and agentic capabilities. It supports tool use and demonstrates competitive reasoning... (description from the OpenRouter listing)

in: imagein: textin: fileout: textReasoningTool useStructured output

Epoch Capabilities Index

145.6 #77 of 268

90% CI 143.2 – 147.3

Listing ↗

Released

16 Apr 2025

Context window

200K tokens

Max output

100K tokens

Input $ / 1M tokens

$1.10

Output $ / 1M tokens

$4.40

Cached input $ / 1M

$0.28

Each benchmark shown separately with its own source

Benchmark results (21)

All benchmarks

Benchmark	Domain	Score	vs best recorded	Setting	Run	Source
LMCA	agents	26.5%	40%	high	—	External ↗
DTBench	reasoning	77.6%	79%	high	—	External ↗
Mystery Game Puzzles	games	5.0% ±2.2	6%	high	27 Aug 2026	Epoch ↗
FrontierMath-Tiers-1-3-v2-Private	math	36.1% ±2.9	39%	high	11 Jun 2026	Eval log ↗
FrontierMath-Tier-4-v2-Private	math	4.9% ±3.4	5%	high	11 Jun 2026	Eval log ↗
Chess Puzzles	games	26.0% ±4.4	36%	high	8 Dec 2025	Eval log ↗
GDPval	agents	25.3%	51%	high	—	External ↗
SimpleQA Verified	knowledge	18.8% ±1.2	25%	high	27 Aug 2026	Eval log ↗
FrontierMath-Tier-4-2025-07-01-Privatesuperseded	math	6.3% ±3.5	13%	high	1 Jul 2025	Epoch ↗
GSO-Bench	coding	3.6%	8%	high	—	External ↗
ARC-AGI-2	reasoning	6.1%	6%	high	—	External ↗
GeoBench	multimodal	64.0%	73%	high	—	External ↗
FrontierMath-2025-02-28-Privatesuperseded	math	24.8% ±2.5	47%	high	13 Nov 2025	Epoch ↗
HLE	knowledge	18.1%	33%	high	—	External ↗
WeirdML	coding	52.6% ±0.0	56%	high	—	External ↗
Aider polyglot	coding	72.0%	82%	high	—	External ↗
OTIS Mock AIME 2024-2025	math	81.7% ±4.7	82%	high	16 Apr 2025	Eval log ↗
SimpleBench	reasoning	38.7%	47%	high	—	External ↗
GPQA diamond	science	79.6% ±2.4	83%	high	16 Apr 2025	Eval log ↗
MATH level 5	math	97.8% ±0.3	100%	high	16 Apr 2025	Epoch ↗
ARC-AGI	reasoning	58.7%	60%	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

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