modelbenchmark.io

GPT-4o mini

OpenAI · gpt-4o-mini

Compare

Small omni GPT for cheap multimodal assistance and production-scale traffic

Specification

openai listing

Context
128K
Max output
16K
Released
2024-07-18
Knowledge cutoff
2023-09
Retires
2027-01-14
Open weights
no
Input
text, image, pdf
Output
text

Price

US dollars per million tokens · openai list price

Input
$0.15
Output
$0.60
Cache read
$0.075
Cache write
$0.15

Batch and priority prices

US dollars per million tokens · per listing

PriceListing$/M
Batch inputazure$0.083
Batch inputazure$0.075
Batch outputazure$0.30
Priority inputopenai$0.25
Priority outputopenai$1.00

Batch input is 45% below the interactive input price at the same listing.

Quality

8 benchmarks · 2 sources

Composite
13th
percentile of 344 scored models
Rank
300 / 344
±0.05 sd
Evidence
8 × 2
3 or more benchmarks, from 2 or more sources
Effort range
—
one configuration only
coding4th1/3 bench
reasoning19th3/4 bench
math15th3/7 bench
knowledge1st1/1 bench

Every score

one row per source and configuration — nothing averaged away

BenchmarkScoreConfigurationSourceRun
MATH level 552.6 ±1.1defaultEpoch AI2025-01-27
Mystery Game Puzzles12.0 ±3.3defaultEpoch AI2026-08-27
GPQA diamond37.7 ±2.4defaultEpoch AI2025-01-27
Chess Puzzles3.0 ±1.7defaultEpoch AI2026-10-13
OTIS Mock AIME 2024-20256.9 ±2.7defaultEpoch AI2025-07-30
FrontierMath-Tiers-1-3-v20.7 ±0.5defaultEpoch AI2026-08-27
Aider polyglot3.6default · aider wholeAider2024-12-21
SimpleQA Verified8.3 ±0.9defaultEpoch AI2026-08-31

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

Available from 27 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
openaivendor$0.15$0.60$0.075—128K16K2027-01-14
cloudflare-ai-gateway$0.075$0.30$0.0375—128K16K—
poe$0.1364$0.5455$0.0682—124K4K—
abacus$0.15$0.60——128K16K—
aimlapi$0.15$0.60$0.075—128K16K—
azure
$0.15
$0.165
$0.60
$0.66
$0.075
$0.083
—128K16K2027-04-14
azure-cognitive-services$0.15$0.60$0.075—128K16K—
crossmodel$0.15$0.60$0.075$0.15128K16K—
edenai$0.15$0.60$0.075—128K16K—
gmi · gmi$0.15$0.60——131K16K—
helicone$0.15$0.60$0.075—128K16K—
impossibl$0.15$0.60$0.075—128K16K—
kilo$0.15$0.60$0.075—128K16K—
llmgateway$0.15$0.60$0.075—128K16K—
llmgateway-providers$0.15$0.60$0.075—128K16K—
lowrouter · auto$0.15$0.60$0.075—128K16K—
merge-gateway$0.15$0.60$0.075—128K16K—
nano-gpt$0.15$0.60$0.075—128K16K—
ofox$0.15$0.60$0.075—128K16K—
openrouter$0.15$0.60$0.075—128K16K—
orcarouter$0.15$0.60$0.075—128K16K—
prime$0.15$0.60——128K16K—
replicate · replicate$0.15$0.60—————
vercel$0.15$0.60$0.075—128K16K—
vercel_ai_gateway · vercel-ai-gateway$0.15$0.60$0.075free128K16K—
cortecs$0.159$0.638$0.081—128K16K—
github_copilot · github-copilot————64K4K—

Price history

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

DateHostField$/M
2026-09-02ofoxCache read$0.075
2026-09-19ofoxCache read$0.06
2026-10-17ofoxCache read$0.075
2026-09-02openrouterCache read$0.075
2026-09-06openrouterCache read$0.075
2026-10-10openrouterCache read$0.075
2026-09-02poeCache read$0.068
2026-10-22poeCache read$0.0682
2026-09-02ofoxInput$0.15
2026-09-19ofoxInput$0.12
2026-10-17ofoxInput$0.15
2026-09-02openrouterInput$0.15
2026-09-06openrouterInput$0.15
2026-10-10openrouterInput$0.15
2026-09-02poeInput$0.14
2026-10-22poeInput$0.1364
2026-09-02ofoxOutput$0.60
2026-09-19ofoxOutput$0.48
2026-10-17ofoxOutput$0.60
2026-09-02openrouterOutput$0.60
2026-09-06openrouterOutput$0.60
2026-10-10openrouterOutput$0.60
2026-09-02poeOutput$0.54
2026-10-22poeOutput$0.5455
2026-10-13aimlapiCache read$0.0975
2026-10-23aimlapiCache read$0.075
2026-10-13aimlapiInput$0.195
2026-10-23aimlapiInput$0.15
2026-10-13aimlapiOutput$0.78
2026-10-23aimlapiOutput$0.60

More from OpenAI

most-hosted first