All Releases••

Proprietary release

•

Released September 2026

GPT-6 Astra: Release Intelligence, Performance & Price

Model summary

IntelligenceUpdated

57
Artificial Analysis Intelligence Index
4 out of 4 units for Intelligence.

Performance

Time to first answer token 228.86sOutput speed 36 tok/s
24.2
Time per Intelligence Index task (minutes)
4 out of 4 units for Performance.

Cost

In $10.00Out $50.00Cache discount 90%
$6.24
Cost per Intelligence Index task
4 out of 4 units for Cost.

Verbosity

Input tokens 1.68BOutput/input 3.232%
54.3M
Output tokens from Intelligence Index
2 out of 4 units for Verbosity.

The GPT-6 Astra release offers 5 models, each with different intelligence, performance, and pricing characteristics. Below is a comparison of the key metrics across the 5 models.

  • For intelligence, the top model of GPT-6 Astra is GPT-6 Astra (Max) at 57.
  • For speed, GPT-6 Astra (Low) at 4.0 min has the lowest time per task.
  • For latency, GPT-6 Astra (Low) at 3.64s offers the lowest time to first answer token.
  • For pricing, GPT-6 Astra (Low) at $1.42 offers the lowest cost per task. Prices vary up to 4.4x across models.

Updated
Artificial Analysis Intelligence Index · Higher is better
No publicly available API
Inference time (minutes) per Intelligence Index task · Lower is better
No publicly available API
Weighted average cost (USD) per Intelligence Index task · Lower is better
No publicly available API

IntelligenceUpdated

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v5.0.1 incorporates 9 evaluations: AA-Briefcase v1.1, AutomationBench-AA, Terminal-Bench 4.0, FrontierCode-AA 2.0, AA-Omniscience v1.1, GDP.pdf, HLE-Diamond, CritPt v2 (Hard Subset), and Terminal-Bench-Science 0.1

Intelligence Index vs. Cost per Task

Artificial Analysis Intelligence Index · Weighted average cost (USD) per Artificial Analysis Intelligence Index task
Most attractive quadrant
GPT-6 Astra
Pareto line

Cost

Cost per Intelligence Index Task

Weighted average cost (USD) per Intelligence Index task · Segmented by token type · Lower is better

Speed & Latency

Time per Intelligence Index Task

Output tokens divided by end-to-end output speed (minutes per Intelligence Index task) · Speed at the model's own endpoint or the median across providers · Lower is better

Capability Scores

Capability Indexes

Measures the performance of models on specific capabilities and industries
Finance & Accounting Index

Incorporates 5 evaluations: AA-Briefcase v1.1, AA-Omniscience v1.1, HLE-Diamond, AutomationBench-AA, GDP.pdf · Higher is better

Strategy & Ops Index

Incorporates 4 evaluations: AutomationBench-AA, AA-Briefcase v1.1, AA-Omniscience v1.1, GDP.pdf · Higher is better

Legal Index

Incorporates 5 evaluations: AA-Omniscience v1.1, AA-Briefcase v1.1, HLE-Diamond, GDP.pdf, AutomationBench-AA · Higher is better

Healthcare & Medical Index

Incorporates 5 evaluations: AA-Omniscience v1.1, AA-Briefcase v1.1, MLCR-AA, AutomationBench-AA, HLE-Diamond · Higher is better

Engineering Index

Incorporates 7 evaluations: AA-Omniscience v1.1, CritPt v2 (Hard Subset), Terminal-Bench-Science 0.1, HLE-Diamond, Terminal-Bench 4.0, FrontierCode-AA 2.0, AA-Briefcase v1.1 · Higher is better

Economics Index

Incorporates 4 evaluations: AA-Omniscience v1.1, AA-Briefcase v1.1, HLE-Diamond, GDP.pdf · Higher is better

Evaluations

Evaluations measured independently by Artificial Analysis · AA-PersonalityBench scores are percentile ranks from 1 to 100
See more

Agentic knowledge work, Elo

1 - hallucination rate

Agentic SaaS workflows

Agentic scientific research workflows in a terminal

Agentic software engineering in open-source repositories

Reasoning & knowledge

Professional document reasoning, All-pass

Agentic coding & terminal use

Physics reasoning

Quantitative analysis on spreadsheets & documents

Legal agentic work, Hallucination-Gated All-Pass Rate

Agentic real-world work tasks, Elo

Agentic interactive 3D experiences, Elo

Product and brand videos designed in code, Elo

Kubernetes incident root-cause analysis

MLCR-AAUpdated

Medical long context reasoning

Visual reasoning

Rich, varied, well-crafted language

Well-structured, scannable, answer-first

Truthful, acknowledges uncertainty, non-sycophantic

Use of filler and generic language

AA-PersonalityBenchNew

Personality Indexes

Each axis is one of 8 personality indexes: the mean of its traits’ Elo z-scores, percentile ranked from 1 to 100 across all models evaluated in the benchmark

More from OpenAI

Further details

Weights
Provider Benchmarks
GPT-6 Astra (Max)
OpenAI logoOpenAI
57
-
1M
$7.70
24.2
36
Not available
OpenAIMicrosoft AzureAmazon Bedrock
GPT-6 Astra (Xhigh)
OpenAI logoOpenAI
55
-
1M
$7.70
11.7
38
Not available
OpenAI
GPT-6 Astra (High)
OpenAI logoOpenAI
55
-
1M
$7.70
9.6
40
Not available
OpenAI
GPT-6 Astra (Medium)
OpenAI logoOpenAI
52
-
1M
$7.70
7.5
40
Not available
OpenAI
GPT-6 Astra (Low)
OpenAI logoOpenAI
43
-
1M
$7.70
4.0
38
Not available
OpenAIMicrosoft Azure