modelbenchmark.io

GPT-6 Luna

OpenAI · gpt-6-luna

Compare

OpenAI's most efficient model for focused, high-volume tasks

Specification

openai listing

Context
1.1M
Max output
128K
Released
2026-09-22
Knowledge cutoff
2026-05-18
Retires
2028-03-11
Open weights
no
Input
text, image, pdf
Output
text

Price

US dollars per million tokens · openai list price

Input
$0.10
Output
$0.50
Cache read
$0.01
Cache write
$0.125

Batch and priority prices

US dollars per million tokens · per listing

PriceListing$/M
Batch inputopenai$0.05
Batch outputopenai$0.25
Priority inputopenai$0.20
Priority outputopenai$1.00
Priority outputdatabricks$1.0003

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 creation input token cost above 12hr$0.50
272KCache read$0.02
272KCache write$0.25
272KInput$0.20
272KOutput$0.75

Quality

12 benchmarks · 2 sources

Composite
60th
percentile of 344 scored models
Rank
138 / 344
±0.07 sd
Evidence
12 × 2
3 or more benchmarks, from 2 or more sources
Effort range
—
one configuration only
reasoning65th3/4 bench
math90th3/7 bench
knowledge48th1/1 bench
general24th1/1 bench

Every score

one row per source and configuration — nothing averaged away

BenchmarkScoreConfigurationSourceRun
OTIS Mock AIME 2024-202598.9 ±0.7maxEpoch AI2026-09-22
FrontierMath-Tiers-1-3-v279.0 ±2.4maxEpoch AI2026-09-22
GPQA diamond90.5 ±1.9maxEpoch AI2026-09-22
FrontierMath-Tier-4-v256.1 ±7.8maxEpoch AI2026-09-22
Chess Puzzles36.0 ±4.8maxEpoch AI2026-10-13
Furniture Assembly44.2 ±6.4maxEpoch AI2026-09-28
SimpleQA Verified41.4 ±1.6maxEpoch AI2026-09-22
LiveBench72.2max · 23 runsLiveBench2026-06-25
SpireClimb (Slay the Spire 1, Ascension 20 Heart)0.0 ±0.0maxEpoch AI2026-09-28
SpireClimb0.0 ±0.0maxEpoch AI2026-09-28
Mystery Game Puzzles7.0 ±2.6maxEpoch AI2026-09-22
SpireClimb (Slay the Spire 2, Ascension 10)0.0 ±0.0maxEpoch AI2026-09-28

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

Available from 63 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
openaivendor$0.10$0.50$0.01$0.125
1.1M
922K
128K—
kunavo$0.04$0.20$0.004$0.051.1M128K—
outerlabs$0.06$0.30$0.006—1.1M128K—
uplinkswitch$0.0625$0.20$0.0125—1.1M128K—
iteracompute$0.08$0.40$0.008$0.101.1M128K—
302ai$0.10$0.50$0.01$0.1251.1M128K—
aihubmix$0.10$0.50$0.01$0.1251.1M128K—
aimlapi$0.10$0.50$0.01$0.1251.1M128K—
aixy$0.10$0.50$0.01$0.1251.1M128K—
amazon-bedrock · global$0.10$0.50$0.01$0.1251.1M128K—
azure
$0.10
$0.11
$0.12
$0.50
$0.55
$0.60
$0.01
$0.011
$0.012
$0.125
$0.1375
$0.15
1.1M
922K
128K2028-03-11
azure_ai$0.10$0.50$0.01$0.125922K128K—
azure-cognitive-services$0.10$0.50$0.01$0.1251.1M128K—
bedrock_converse$0.10$0.50$0.01$0.125922K128K—
bedrock_converse · global$0.10$0.50$0.01$0.125922K128K—
boundless$0.10$0.50$0.01—1.1M128K—
cheapest-token$0.10$0.50$0.01$0.1251.1M128K—
cloudflare-ai-gateway$0.10$0.50$0.01$0.1251.1M128K—
commandcode$0.10$0.50$0.01$0.1251.1M128K—
crossmodel$0.10$0.50$0.01$0.1251.1M128K—
databricks$0.10$0.50$0.01$0.125922K128K—
edenai$0.10$0.50$0.01$0.1251.1M128K—
fastrouter$0.10$0.50$0.01$0.125400K128K—
frogbot$0.10$0.50$0.01$0.1251.1M128K—
github_copilot · github-copilot$0.10$0.50$0.01$0.125922K128K—
github-copilot$0.10$0.50$0.01$0.1251.1M128K—
gmi · gmi$0.10$0.50$0.01—1.1M——
kilo$0.10$0.50$0.01$0.1251.1M128K—
llmgateway$0.10$0.50$0.01$0.1251.1M128K—
llmgateway-providers$0.10$0.50$0.01$0.1251.1M128K—
lowrouter · auto$0.10$0.50$0.01$0.1251.1M128K—
merge-gateway$0.10$0.50$0.01$0.1251.1M128K—
nano-gpt$0.10$0.50$0.01$0.1251.1M128K—
nearai$0.10$0.50$0.01—1.1M128K—
neon$0.10$0.50$0.01$0.1251.1M128K—
ofox$0.10$0.50$0.01$0.1251.1M128K—
opencode$0.10$0.50$0.01$0.1251.1M128K—
opencode_go · opencode-go$0.10$0.50$0.01$0.125922K128K—
opencode-go$0.10$0.50$0.01$0.1251.1M128K—
openrouter$0.10$0.50$0.01$0.1251.1M128K—
opper$0.10$0.50$0.01$0.1251.1M128K—
perplexity$0.10$0.50$0.01————
perplexity-agent$0.10$0.50$0.01$0.1251.1M128K—
prime$0.10$0.50——1.1M128K—
requesty$0.10$0.50$0.01$0.1251.1M128K—
sparrowcodes$0.10$0.50$0.01—1.1M128K—
standardcompute$0.10$0.50$0.01$0.1251.1M50K—
synthorai$0.10$0.50$0.01—1.1M128K—
vercel$0.10$0.50$0.01$0.1251.1M128K—
vivgrid$0.10$0.50$0.01$0.1251.1M128K—
zerogpu · zerogpu$0.10$0.50$0.01—272K128K—
bumo$0.106$0.53$0.0106$0.13251.1M128K—
amazon-bedrock$0.11$0.55$0.011$0.13751.1M128K—
amazon-bedrock · us$0.11$0.55$0.011$0.13751.1M128K—
bedrock_converse · us$0.11$0.55$0.011$0.1375922K128K—
bedrock_mantle · bedrock-mantle$0.11$0.55$0.011$0.1375922K128K—
obscurify$0.1105$0.5525$0.0111$0.13811.1M128K—
cortecs$0.12$0.60$0.012$0.151.1M128K—
dothoosh$0.1283$0.6417$0.0128$0.16041.1M128K—
void-api$0.132$0.66$0.0132$0.1381.1M128K—
kyma · kyma$0.1639$0.8195$0.0135—1.1M128K—
chatgpt · chatgpt————922K128K—
sap-ai-core————1.1M128K—

Price history

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

DateHostField$/M
2026-09-23azureCache read$0.01
2026-09-24azureCache read$0.01
2026-09-23nano-gptCache read$0.005
2026-09-24nano-gptCache read$0.01
2026-09-23openrouterCache read$0.01
2026-10-10openrouterCache read$0.01
2026-09-23azureCache write$0.125
2026-09-24azureCache write$0.125
2026-09-23nano-gptCache write$0.0625
2026-09-24nano-gptCache write$0.125
2026-09-23openrouterCache write$0.125
2026-10-10openrouterCache write$0.125
2026-09-23azureInput$0.10
2026-09-24azureInput$0.10
2026-09-23nano-gptInput$0.05
2026-09-24nano-gptInput$0.10
2026-09-23openrouterInput$0.10
2026-10-10openrouterInput$0.10
2026-09-23azureOutput$0.50
2026-09-24azureOutput$0.50
2026-09-23nano-gptOutput$0.25
2026-09-24nano-gptOutput$0.50
2026-09-23openrouterOutput$0.50
2026-10-10openrouterOutput$0.50
2026-09-24ofoxCache read$0.008
2026-10-17ofoxCache read$0.01
2026-09-24ofoxCache write$0.10
2026-10-17ofoxCache write$0.125
2026-09-24ofoxInput$0.08
2026-10-17ofoxInput$0.10
2026-09-24ofoxOutput$0.40
2026-10-17ofoxOutput$0.50
2026-10-13aimlapiCache read$0.013
2026-10-23aimlapiCache read$0.01
2026-10-13aimlapiInput$0.13
2026-10-23aimlapiInput$0.10
2026-10-13aimlapiOutput$0.65
2026-10-23aimlapiOutput$0.50
2026-10-21aimlapiCache write$0.1625
2026-10-23aimlapiCache write$0.125

Other listings of this model

same model, different key

More from OpenAI

most-hosted first