modelbenchmark.io

GPT-4o (2024-05-13)

OpenAI · gpt-4o

Compare

GPT model for general reasoning, writing, coding, and tool-assisted tasks

Specification

openai listing

Context
128K
Max output
4K
Released
2024-05-13
Knowledge cutoff
2023-09
Retires
2026-10-23
Open weights
no
Input
text, image
Output
text

Price

US dollars per million tokens · openai list price

Input
$5.00
Output
$15.00
Cache read
$1.25
Cache write
$2.50

Batch and priority prices

US dollars per million tokens · per listing

PriceListing$/M
Batch inputopenai$1.25
Batch inputopenai$2.50
Batch outputopenai$5.00
Batch outputopenai$7.50
Priority inputopenai$4.25
Priority inputopenai$8.75
Priority outputopenai$17.00
Priority outputopenai$26.25

Batch input is 75% below the interactive input price at the same listing.

Quality

9 benchmarks · 3 sources

Composite
21st
percentile of 334 scored models
Rank
263 / 334
±0.04 sd
Evidence
9 × 3
3 or more benchmarks, from 2 or more sources
Effort range
2.0
points between effort settings
coding15th2/3 bench
reasoning41st2/4 bench
math16th4/7 bench
knowledge23rd1/1 bench

Every score

one row per source and configuration — nothing averaged away

BenchmarkScoreConfigurationSourceRun
MATH level 551.4 ±1.1default · 3 runsEpoch AI2025-02-05
Chess Puzzles13.0 ±3.4defaultEpoch AI2026-07-15
GPQA diamond48.7 ±2.6default · 3 runsEpoch AI2025-02-05
SimpleQA Verified26.0 ±1.4defaultEpoch AI2026-08-31
FrontierMath-v10.2 ±0.3default · 4 runsEpoch AI2025-03-07
Aider polyglot20.6default · aider diff · 2 runsAider2024-12-30
SWE-Bench verified31.0 ±2.1defaultEpoch AI2026-02-11
29.0default · MASAI · 8 runsSWE-bench
Sources and settings disagree by 2.0 points. Both figures stand.
OTIS Mock AIME 2024-20256.3 ±2.6default · 3 runsEpoch AI2025-02-25
FrontierMath-Tiers-1-3-v20.3 ±0.4defaultEpoch 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 26 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
openaivendor
$5.00
$2.50
$15.00
$10.00
$1.25128K
4K
16K
2026-10-23
cloudflare-ai-gateway$1.25$5.00$0.625128K16K
302ai$2.50$10.00128K16K
abacus$2.50$10.00$1.25128K16K
azure
$2.50
$2.75
$5.00
$10.00
$11.00
$15.00
$1.25
$1.375
$1.38128K
16K
4K
2027-04-14
2026-10-01
azure-cognitive-services$2.50$10.00$1.25128K16K
edenai$2.50$10.00$1.25128K16K
frogbot$2.50$10.00$1.25128K16K
gmi · gmi$2.50$10.00131K16K
helicone$2.50$10.00$1.25128K16K
impossibl$2.50$10.00$1.25128K16K
llmgateway$2.50$10.00$1.25128K16K
llmgateway-providers$2.50$10.00$1.25128K16K
nano-gpt$2.50$10.00$1.25128K16K
ofox$2.50$10.00$1.25128K16K
pioneer$2.50$10.00$1.25$2.50128K16K
replicate · replicate$2.50$10.00
vercel$2.50$10.00$1.25128K16K
vercel_ai_gateway · vercel-ai-gateway$2.50$10.00$1.25free128K16K
cortecs$2.659$10.635$1.33128K16K
kilo
$5.00
$2.50
$15.00
$10.00
$1.25128K
4K
16K
merge-gateway
$5.00
$2.50
$15.00
$10.00
$1.25128K
4K
16K
openrouter
$5.00
$2.50
$15.00
$10.00
$1.25128K
4K
16K
orcarouter
$5.00
$2.50
$15.00
$10.00
$1.25128K
4K
16K
github_copilot · github-copilot64K
4K
16K
poe128K8K

Price history

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

DateHostField$/M
2026-09-02openrouterCache read$1.25
2026-09-06openrouterCache read$1.25
2026-09-02kiloInput$2.50
2026-09-04kiloInput$5.00
2026-09-02merge-gatewayInput$2.50
2026-09-04merge-gatewayInput$5.00
2026-09-02openaiInput$2.50
2026-09-04openaiInput$5.00
2026-09-02openrouterInput$2.50
2026-09-04openrouterInput$5.00
2026-09-02orcarouterInput$2.50
2026-09-04orcarouterInput$5.00
2026-09-02kiloOutput$10.00
2026-09-04kiloOutput$15.00
2026-09-02merge-gatewayOutput$10.00
2026-09-04merge-gatewayOutput$15.00
2026-09-02openaiOutput$10.00
2026-09-04openaiOutput$15.00
2026-09-02openrouterOutput$10.00
2026-09-04openrouterOutput$15.00
2026-09-02orcarouterOutput$10.00
2026-09-04orcarouterOutput$15.00

More from OpenAI

most-hosted first