modelbenchmark.io

GPT-5.4

OpenAI · gpt-5-4

Compare

Agent-ready GPT for coding and computer-use workflows at a lower cost

Specification

openai listing

Context
1.1M
Max output
128K
Released
2026-03-05
Knowledge cutoff
2025-08-31
Retires
2027-09-02
Open weights
no
Input
text, image, pdf
Output
text

Price

US dollars per million tokens · openai list price

Input
$2.50
Output
$15.00
Cache read
$0.25
Cache write
free

Batch and priority prices

US dollars per million tokens · per listing

PriceListing$/M
Batch inputazure$1.25
Batch inputazure$1.375
Batch outputazure$7.50
Batch outputazure$8.25
Priority inputazure_ai$5.00
Priority inputazure$5.50
Priority outputazure_ai$30.00
Priority outputazure$33.00

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

Long-context price tiers

the rate above each context threshold

AbovePrice$/M
272KCache read$0.50
272KInput$5.00
272KOutput$22.50

Quality

14 benchmarks · 2 sources

Composite
85th
percentile of 344 scored models
Rank
53 / 344
±0.06 sd
Evidence
14 × 2
3 or more benchmarks, from 2 or more sources
Effort range
16.7
points between effort settings
coding47th2/3 bench
reasoning75th4/4 bench
math93rd5/7 bench
knowledge52nd1/1 bench
general73rd1/1 bench

By reasoning effort

every setting placed on the same scale as the leaderboard

SettingCompositeBenchmarks behind it
high95th7
xhigh87th12
medium86th4
low66th4

Every score

one row per source and configuration — nothing averaged away

BenchmarkScoreConfigurationSourceRun
FrontierMath-Tier-4-2025-07-0150.0 ±50.0highEpoch AI2026-04-02
↳38.5 ±50.0xhigh · 2 runsEpoch AI2026-04-02
Settings differ by 11.5 points on this benchmark.
FrontierMath-v180.0 ±13.3highEpoch AI2026-04-02
↳47.6 ±2.9xhighEpoch AI2026-03-06
Settings differ by 32.4 points on this benchmark.
FrontierMath-Tiers-1-3-v278.6 ±2.4xhighEpoch AI2026-06-11
OTIS Mock AIME 2024-202597.8 ±2.2highEpoch AI2026-07-15
↳95.6 ±3.1mediumEpoch AI2026-07-15
↳95.3 ±3.2xhighEpoch AI2026-03-06
↳84.4 ±5.5lowEpoch AI2026-07-15
Settings differ by 13.4 points on this benchmark.
GPQA diamond93.3 ±1.8xhighEpoch AI2026-03-06
↳89.9 ±2.1highEpoch AI2026-07-15
↳88.9 ±2.2mediumEpoch AI2026-07-15
↳84.8 ±2.6lowEpoch AI2026-07-15
Settings differ by 8.5 points on this benchmark.
SWE-Bench verified76.9 ±1.9highEpoch AI2026-03-06
LiveBench78.8xhigh · 23 runsLiveBench2026-06-25
Chess Puzzles48.0 ±5.0xhighEpoch AI2026-10-13
↳42.0 ±5.0highEpoch AI2026-10-13
↳37.0 ±4.9mediumEpoch AI2026-10-13
↳19.0 ±3.9lowEpoch AI2026-10-13
Settings differ by 29.0 points on this benchmark.
FrontierMath-Tier-4-v249.0 ±8.0xhighEpoch AI2026-06-11
Mystery Game Puzzles37.0 ±4.9xhighEpoch AI2026-07-24
↳28.0 ±4.5mediumEpoch AI2026-08-28
↳17.0 ±3.8lowEpoch AI2026-08-27
Settings differ by 20.0 points on this benchmark.
SimpleQA Verified45.1 ±1.6xhighEpoch AI2026-08-27
EBR-bench25.4xhighEpoch AI2026-06-25
Furniture Assembly37.5 ±6.2xhighEpoch AI2026-09-10
MirrorCode15.6 ±7.7highEpoch AI2026-08-10

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

Available from 52 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
openaivendor$2.50$15.00$0.25—1.1M128K—
unorouter · freefreefree——1.1M128K—
xpersona$0.75$6.00$0.075—1.1M128K—
unorouter$1.80$10.80——1.1M128K—
poe$2.2727$13.6364$0.2273—1.1M128K—
302ai$2.50$15.00$0.25free1.1M128K—
abacus$2.50$15.00$0.25—400K128K—
ai-router$2.50$15.00$0.25—1.1M128K—
aihubmix$2.50$15.00$0.25—1.1M128K—
aimlapi$2.50$15.00$0.25—1.1M128K—
azure
$2.50
$2.75
$15.00
$16.50
$0.25
$0.275
—
1.1M
922K
128K2027-09-02
azure_ai$2.50$15.00$0.25—922K128K2027-09-02
azure-cognitive-services$2.50$15.00$0.25—1.1M128K—
bedrock_converse · global$2.50$15.00$0.25—922K128K—
cloudflare-ai-gateway$2.50$15.00$0.25—1M128K—
commandcode$2.50$15.00$0.25free400K128K—
crossmodel$2.50$15.00$0.25$2.501.1M128K—
daoxe$2.50$15.00$0.25—1.1M128K—
databricks$2.50$15.00$0.25$2.50
1.1M
922K
128K—
edenai$2.50$15.00$0.25—1.1M128K—
freemodel$2.50$15.00$0.25$2.501.1M128K—
github_copilot · github-copilot$2.50$15.00$0.25—272K128K2026-10-19
github-copilot$2.50$15.00$0.25—1.1M128K—
impossibl$2.50$15.00$0.25—1.1M128K—
kilo$2.50$15.00$0.25—1.1M128K—
llmgateway$2.50$15.00$0.25—1.1M128K—
llmgateway-providers$2.50$15.00$0.25—1.1M128K—
lowrouter · auto$2.50$15.00$0.25—1.1M128K—
merge-gateway$2.50$15.00$0.25—1.1M128K—
nano-gpt$2.50$15.00$0.25—1.1M128K—
nearai$2.50$15.00$0.25—1.1M128K—
neon$2.50$15.00$0.25—1.1M128K—
ofox$2.50$15.00$0.25—1.1M128K—
opencode$2.50$15.00$0.25—1.1M128K—
openrouter$2.50$15.00$0.25—1.1M128K—
opper$2.50$15.00$0.25—1.1M128K—
orcarouter$2.50$15.00$0.25—1.1M128K—
perplexity$2.50$15.00$0.25————
perplexity-agent$2.50$15.00$0.25—1.1M128K—
prime$2.50$15.00——1.1M128K—
sap-ai-core$2.50$15.00$0.25—1.1M128K—
vercel$2.50$15.00$0.25—1.1M128K—
vivgrid$2.50$15.00$0.25—400K128K—
zenmux$2.50$15.00$0.25—1.1M128K—
amazon-bedrock$2.75$16.50$0.275—1M128K—
bedrock_converse · us$2.75$16.50$0.275—922K128K—
bedrock_mantle · bedrock-mantle
$2.75
$3.375
$3.30
$16.50
$20.25
$19.80
$0.275
$0.3375
$0.33
—922K128K—
requesty$2.75$16.50$0.275—1.1M128K—
cortecs$2.898$15.453$0.242—1.1M128K—
anyapi————1.1M128K—
chatgpt · chatgpt————1.1M128K—
model-oracle-ai————1.1M128K—

Price history

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

DateHostField$/M
2026-09-02azureCache read$0.25
2026-09-17azureCache read$0.25
2026-09-02bedrock_mantle · bedrock-mantleCache read$0.275
2026-09-04bedrock_mantle · bedrock-mantleCache read$0.275
2026-10-08bedrock_mantle · bedrock-mantleCache read$0.275
2026-09-02ofoxCache read$0.25
2026-09-19ofoxCache read$0.20
2026-10-17ofoxCache read$0.25
2026-09-02openrouterCache read$0.25
2026-09-06openrouterCache read$0.25
2026-10-10openrouterCache read$0.25
2026-09-02poeCache read$0.22
2026-10-22poeCache read$0.2273
2026-09-02bedrock_mantle · bedrock-mantleInput$2.75
2026-09-04bedrock_mantle · bedrock-mantleInput$2.75
2026-10-08bedrock_mantle · bedrock-mantleInput$2.75
2026-09-02ofoxInput$2.50
2026-09-19ofoxInput$2.00
2026-10-17ofoxInput$2.50
2026-09-02openrouterInput$2.50
2026-09-06openrouterInput$2.50
2026-10-10openrouterInput$2.50
2026-09-02poeInput$2.20
2026-10-22poeInput$2.2727
2026-09-02zenmuxInput$3.75
2026-10-22zenmuxInput$2.50
2026-09-02bedrock_mantle · bedrock-mantleOutput$16.50
2026-09-04bedrock_mantle · bedrock-mantleOutput$16.50
2026-10-08bedrock_mantle · bedrock-mantleOutput$16.50
2026-09-02ofoxOutput$15.00
2026-09-19ofoxOutput$12.00
2026-10-17ofoxOutput$15.00
2026-09-02openrouterOutput$15.00
2026-09-06openrouterOutput$15.00
2026-10-10openrouterOutput$15.00
2026-09-02poeOutput$14.00
2026-10-22poeOutput$13.6364
2026-09-02zenmuxOutput$18.75
2026-10-22zenmuxOutput$15.00
2026-10-13aimlapiCache read$0.325
2026-10-23aimlapiCache read$0.25
2026-10-13aimlapiInput$3.25
2026-10-23aimlapiInput$2.50
2026-10-13aimlapiOutput$19.50
2026-10-23aimlapiOutput$15.00

Other listings of this model

same model, different key

More from OpenAI

most-hosted first