Advertise here
#21

GPT-6 Astra

OpenAI Release: 2026-09-04 Tested on: 2026-10-01 04:32 openai/gpt-6-astra::medium
~2T total (~150B active)MoEClosedEstimated
(max) (xhigh) (high) (medium) (low)

Summary

GPT-6 Astra scores 9.5 on AI BENCHY and ranks #21. It has 9.9 reliability, a 91.3% pass rate, $2.707 total cost, and 9.90s average response time.

Model facts

Researched on 2026-09-04

Estimated
Parameters
~2T total (~150B active)
Architecture
MoE
Availability
Closed
License
-

Best estimate from public evidence; the vendor did not disclose every value. OpenAI does not disclose parameter counts or architecture. Low-confidence catalog estimate reuses the GPT-5.6 Sol scale (2000B total, 150B active); this is not a reported Astra specification.

Parameter source Availability source

Score

9.5

Consistency

9.3

Reliability

9.9

Total Cost (Current Price)

$2.707

Total Output Tokens

21,045

Total Input Tokens

165,464

Input Price

$10.000 / 1M

Output Price

$50.000 / 1M

Cache Read Price

$1.000 / 1M

Cache Write Price

$12.500 / 1M

Cache prices apply to input tokens. Reads reuse cached prompts; writes store them and can cost extra. Output tokens use the output price.

Tests Correct

Wrong Tests: 3

Attempt pass rate: 91.3%

Flaky tests

2

Flaky tests had mixed outcomes across runs (at least one pass and one fail).

Response Time (avg)

9.90s

Response Time (max): 73.48s

Response Time (total): 227.67s

Wrong answer: 3

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#21 GPT-6 Astra

medium
Cost
$0.231
Time
81.0s
Tokens
4,708 tok

Price History

Historical pricing data for this model from OpenRouter.

Date Input Price Output Price Cache Read Price Cache Write Price
2026-10-01 18:33 $10.000 / 1M $50.000 / 1M $1.000 / 1M $12.500 / 1M

Charts

Choose the first model, then click a second model to open a side-by-side page.

Total Output Tokens

Score vs Total Output Tokens

Quick Compare

GPT-6 AstramediumvsGPT-5.6 SolmediumGPT-6 AstramediumvsGPT-5.6 SolhighGPT-6 AstramediumvsGPT-6 SolmediumGPT-6 AstramediumvsGPT-5.5lowGPT-6 AstramediumvsGemini 3.7 FlashmediumGPT-6 AstramediumvsClaude Opus 5.5lowGPT-6 AstramediumvsGemini 4 ArgonhighGPT-6 AstramediumvsClaude Opus 5.5highGPT-6 AstramediumvsGPT-6.1 SollowGPT-6 AstramediumvsGemini 3.7 Flashlow

Category Breakdown

Category Score Consistency Tests Correct
Agentic 10.0 10.0
Anti-AI Tricks 10.0 10.0
Coding 10.0 10.0
Combined 10.0 10.0
Data parsing and extraction 10.0 10.0
Domain specific 5.9 7.2
General Intelligence 10.0 10.0
Instructions following 10.0 10.0
Puzzle Solving 10.0 10.0
Tool Calling 10.0 10.0
Trivia 2.8 1.6

Compared models