modelbenchmark.io

o4-mini

OpenAI · o4-mini

Compare

Fast o-series model for compact reasoning, coding, and tool use

Specification

openai listing

Context
200K
Max output
100K
Released
2025-04-16
Knowledge cutoff
2024-05
Retires
2026-10-23
Open weights
no
Input
text, image
Output
text

Price

US dollars per million tokens · openai list price

Input
$1.10
Output
$4.40
Cache read
$0.275
Cache write
free

Batch and priority prices

US dollars per million tokens · per listing

PriceListing$/M
Batch inputazure$0.55
Batch inputazure$0.605
Batch outputazure$2.20
Batch outputazure$2.42
Priority inputopenai$2.00
Priority outputopenai$8.00

Batch input is 50% below the interactive input price at the same listing.

Quality

12 benchmarks · 3 sources

Composite
50th
percentile of 344 scored models
Rank
174 / 344
±0.04 sd
Evidence
12 × 3
3 or more benchmarks, from 2 or more sources
Effort range
—
one configuration only
coding61st2/3 bench
reasoning45th3/4 bench
math51st6/7 bench
knowledge13th1/1 bench

Every score

one row per source and configuration — nothing averaged away

BenchmarkScoreConfigurationSourceRun
MATH level 597.8 ±0.3defaultEpoch AI2025-04-16
Aider polyglot72.0high · aider diffAider2025-04-16
GPQA diamond77.5 ±3.0default · 3 runsEpoch AI2026-08-07
OTIS Mock AIME 2024-202570.9 ±6.7default · 3 runsEpoch AI2026-08-07
FrontierMath-v122.4 ±1.8default · 6 runsEpoch AI2025-11-16
SWE-Bench verified54.8default · PatchPilot-v1.1 · 2 runsSWE-bench—
FrontierMath-Tier-4-2025-07-012.1 ±0.0default · 4 runsEpoch AI2025-08-07
Chess Puzzles16.7 ±3.4default · 3 runsEpoch AI2026-10-13
FrontierMath-Tiers-1-3-v227.0 ±2.2default · 3 runsEpoch AI2026-08-27
Mystery Game Puzzles5.0 ±2.2defaultEpoch AI2026-08-27
FrontierMath-Tier-4-v24.9 ±3.4defaultEpoch AI2026-06-11
SimpleQA Verified19.2 ±1.3default · 2 runsEpoch AI2026-08-27

The leading figure for a benchmark is the median across its configurations, so one heroic high-effort run cannot set the number.

Available from 18 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
openaivendor$1.10$4.40$0.275—200K100K2026-10-23
poe$0.99$4.00$0.25—200K100K—
replicate · replicate$1.00$4.00—————
abacus$1.10$4.40——200K100K—
aihubmix$1.10$4.40$0.275—200K100K—
azure
$1.10
$1.21
$4.40
$4.84
$0.275
$0.303
—200K100K2026-11-19
azure-cognitive-services$1.10$4.40$0.275—200K100K—
cloudflare-ai-gateway$1.10$4.40$0.275—200K100K—
helicone$1.10$4.40$0.275—200K100K—
impossibl$1.10$4.40$0.275—200K100K—
jiekou$1.10$4.40——200K100K—
merge-gateway$1.10$4.40$0.275—200K100K—
nano-gpt$1.10$4.40$0.55—200K100K—
nearai$1.10$4.40$0.275—200K100K—
vercel$1.10$4.40$0.275—200K100K—
vercel_ai_gateway · vercel-ai-gateway$1.10$4.40$0.275free200K100K—
anyapi————200K100K—
model-oracle-ai————200K100K—

Price history

append-only observations · a listing writes a row only when its price moves

DateHostField$/M
2026-09-02azureCache read$0.275
2026-09-17azureCache read$0.275
2026-09-02openrouterCache read$0.275
2026-09-06openrouterCache read$0.275
2026-10-10openrouterCache read$0.275
2026-09-02openrouterInput$1.10
2026-09-06openrouterInput$1.10
2026-10-10openrouterInput$1.10
2026-09-02openrouterOutput$4.40
2026-09-06openrouterOutput$4.40
2026-10-10openrouterOutput$4.40

More from OpenAI

most-hosted first