modelbenchmark.io

GPT-6 Astra

OpenAI · gpt-6-astra

Compare

GPT-6 Astra is OpenAI's most capable model for complex reasoning, coding, computer use, research, and document creation.

Specification

openai listing

Context
1.1M
Max output
128K
Released
2026-09-04
Knowledge cutoff
2026-04-30
Retires
2028-01-11
Open weights
no
Input
text, image, pdf
Output
text

Price

US dollars per million tokens · openai list price

Input
$10.00
Output
$50.00
Cache read
$1.00
Cache write
$12.50

Batch and priority prices

US dollars per million tokens · per listing

PriceListing$/M
Batch inputopenai$5.00
Batch outputopenai$25.00
Priority inputopenai$20.00
Priority outputopenai$100.00
Priority outputdatabricks$99.9999

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$50.00
272KCache read$2.00
272KCache write$25.00
272KInput$20.00
272KOutput$75.00

Quality

15 benchmarks · 2 sources

Composite
99th
percentile of 344 scored models
Rank
3 / 344
±0.07 sd
Evidence
15 × 2
3 or more benchmarks, from 2 or more sources
Effort range
9.8
points between effort settings
coding53rd1/3 bench
reasoning100th4/4 bench
math98th3/7 bench
knowledge99th1/1 bench
general97th1/1 bench

By reasoning effort

every setting placed on the same scale as the leaderboard

SettingCompositeBenchmarks behind it
xhigh100th1
medium100th1
max100th13
low99th1
high95th2

Every score

one row per source and configuration — nothing averaged away

BenchmarkScoreConfigurationSourceRun
Mystery Game Puzzles84.0 ±3.7maxEpoch AI2026-08-30
Chess Puzzles74.0 ±4.4maxEpoch AI2026-10-13
EBR-bench76.2 ±8.4maxEpoch AI2026-08-30
FrontierMath-Tier-4-v297.6 ±2.4highEpoch AI2026-08-30
↳97.6 ±2.4maxEpoch AI2026-08-30
↳97.6 ±2.4xhighEpoch AI2026-08-30
↳97.6 ±2.4mediumEpoch AI2026-08-30
↳87.8 ±5.2lowEpoch AI2026-08-30
Settings differ by 9.8 points on this benchmark.
Furniture Assembly80.0 ±4.9maxEpoch AI2026-09-10
SimpleQA Verified75.6 ±1.4maxEpoch AI2026-08-30
SpireClimb (Slay the Spire 2, Ascension 10)34.7 ±5.5maxEpoch AI2026-09-22
FrontierMath-Tiers-1-3-v293.7 ±1.4maxEpoch AI2026-08-30
SpireClimb25.7 ±3.7maxEpoch AI2026-09-23
SpireClimb (Slay the Spire 1, Ascension 20 Heart)16.7 ±4.9maxEpoch AI2026-09-23
GPQA diamond95.8 ±1.4maxEpoch AI2026-08-30
LiveBench83.0max · 23 runsLiveBench2026-06-25
OTIS Mock AIME 2024-2025100.0 ±0.0maxEpoch AI2026-08-30
FrontierMath-Erdos2.9 ±2.1maxEpoch AI2026-08-28
MirrorCode46.7 ±11.9highEpoch AI2026-08-30

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

Available from 47 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
openaivendor$10.00$50.00$1.00$12.50
1.1M
922K
128K—
302ai$10.00$50.00$1.00—1.1M128K—
aihubmix$10.00$50.00$1.00$12.501.1M128K—
aimlapi$10.00$50.00$1.00$12.501.1M128K—
aixy$10.00$50.00$1.00$12.501.1M128K—
amazon-bedrock · global$10.00$50.00$1.00$12.501.1M128K—
azure
$10.00
$11.00
$12.00
$50.00
$55.00
$60.00
$1.00
$1.10
$1.20
$12.50
$13.75
$15.00
1.1M
922K
128K2028-01-11
azure_ai$10.00$50.00$1.00$12.50922K128K—
azure-cognitive-services$10.00$50.00$1.00$12.501.1M128K—
bedrock_converse · global$10.00$50.00$1.00$12.50922K128K—
cloudflare-ai-gateway$10.00$50.00$1.00$12.001.1M128K—
commandcode$10.00$50.00$1.00$12.501.1M128K—
crossmodel$10.00$50.00$1.00$12.501.1M128K—
databricks$10.00$50.00$1.00$12.50922K128K—
edenai$10.00$50.00$1.00$12.501.1M128K—
github_copilot · github-copilot$10.00$50.00$1.00$12.50922K128K—
github-copilot$10.00$50.00$1.00$12.501.1M128K—
kilo$10.00$50.00$1.00$12.501.1M128K—
llmgateway$10.00$50.00$1.00$12.501.1M1.1M—
llmgateway-providers$10.00$50.00$1.00$12.501.1M128K—
llmgateway-providers · aws-mantle$10.00$50.00$1.00$12.501.1M128K—
lowrouter · auto$10.00$50.00$1.00$12.501.1M128K—
merge-gateway$10.00$50.00$1.00$12.501.1M128K—
nano-gpt$10.00$50.00$1.00$12.501.1M128K—
nearai$10.00$50.00$1.00—1.1M128K—
neon$10.00$50.00$1.00$12.501.1M128K—
ofox$10.00$50.00$1.00$12.501.1M128K—
opencode$10.00$50.00$1.00$12.501.1M128K—
openrouter$10.00$50.00$1.00$12.501.1M128K—
opper$10.00$50.00$1.00$12.501.1M128K—
perplexity$10.00$50.00$1.00$12.50———
perplexity-agent$10.00$50.00$1.00$12.501.1M128K—
prime$10.00$50.00——1.1M128K—
requesty$10.00$50.00$1.00$12.501.1M128K—
standardcompute$10.00$50.00$1.00$12.501.1M50K—
vercel$10.00$50.00$1.00$12.501.1M128K—
vivgrid$10.00$50.00$1.00$12.501.1M128K—
zenmux$10.00$50.00$1.00$12.501.1M128K—
amazon-bedrock$11.00$55.00$1.10$13.751.1M128K—
amazon-bedrock · us$11.00$55.00$1.10$13.751.1M128K—
bedrock_converse · us$11.00$55.00$1.10$13.75922K128K—
bedrock_mantle · bedrock-mantle$11.00$55.00$1.10$13.75922K128K—
cortecs$12.272$61.359$1.228$15.341.1M128K—
infer · official$12.50$62.50$1.25$15.625272K128K—
kyma · kyma$13.50$67.50$1.35—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
2026-09-04openaiCache read$1.00
2026-09-05openaiCache read$1.00
2026-09-04openaiCache write$12.50
2026-09-05openaiCache write$12.50
2026-09-04openaiInput$10.00
2026-09-05openaiInput$10.00
2026-09-04openaiOutput$50.00
2026-09-05openaiOutput$50.00
2026-09-05openrouterCache read$1.00
2026-09-06openrouterCache read$1.00
2026-10-10openrouterCache read$1.00
2026-09-05openrouterCache write$12.50
2026-09-06openrouterCache write$12.50
2026-10-10openrouterCache write$12.50
2026-09-05openrouterInput$10.00
2026-09-06openrouterInput$10.00
2026-10-10openrouterInput$10.00
2026-09-05openrouterOutput$50.00
2026-09-06openrouterOutput$50.00
2026-10-10openrouterOutput$50.00
2026-09-06azureCache read$1.00
2026-09-07azureCache read$1.00
2026-09-30azureCache read$1.00
2026-09-06azureCache write$12.50
2026-09-07azureCache write$12.50
2026-09-30azureCache write$12.50
2026-09-06azureInput$10.00
2026-09-07azureInput$10.00
2026-09-30azureInput$10.00
2026-09-06azureOutput$50.00
2026-09-07azureOutput$50.00
2026-09-30azureOutput$50.00
2026-09-07ofoxCache read$1.00
2026-09-19ofoxCache read$0.80
2026-10-17ofoxCache read$1.00
2026-09-07ofoxCache write$12.50
2026-09-19ofoxCache write$10.00
2026-10-17ofoxCache write$12.50
2026-09-07ofoxInput$10.00
2026-09-19ofoxInput$8.00
2026-10-17ofoxInput$10.00
2026-09-07ofoxOutput$50.00
2026-09-19ofoxOutput$40.00
2026-10-17ofoxOutput$50.00
2026-10-13aimlapiCache read$1.30
2026-10-23aimlapiCache read$1.00
2026-10-13aimlapiInput$13.00
2026-10-23aimlapiInput$10.00
2026-10-13aimlapiOutput$65.00
2026-10-23aimlapiOutput$50.00
2026-10-21aimlapiCache write$16.25
2026-10-23aimlapiCache write$12.50

More from OpenAI

most-hosted first