readable AI benchmarks (simplified)

readable AI benchmarks

OpenAI

GPT-6 Astra (Non-reasoning)

non-reasoning · closed weights · Sep 3, 2026

GPT-6 Astra (Non-reasoning) has a Quality Score of 71.2, ranking 22nd among 284 scored models. Per 1M tokens, pricing is $10.00 input, $50.00 output, $1.00 cached input, and $12.50 cached output. It accepts text and image input, outputs text, and has a 1M token context window. Its Reliability Score is 63.3, ranking 24th among 284 scored models, above average (total average: 35.4). Its Value Score is 39.3, high compared with the total Value average of 29.1.

Quality Score71.2coverage 94.7%: missing CursorBench 3.2 and DeepSWE v1.1
Value Score39.3
Reliability63.3
Cache Discount90%

Model specification

Reasoning
non-reasoning
Input modalities
text, image
Output modalities
text
Context window
1M tokens
Weights
closed weights

Token prices USD per 1M tokens

Input
$10.00
Output
$50.00
Cached Input
$1.00
Cached Output
$12.50
Cost per task
$0.931

Capability

Intelligence
55.3
Coding
76.2
Agentic
48.8
Omniscience
26.6
Correct
55.6%
Blended price
$7.70

Answer outcomes

Attempt rate: 84.7%

Correct 55.6%Incorrect 29.1%Abstained 15.3%

Artificial Analysis benchmarks

GDPval-AA v2
55%
τ³-Banking
37%
Terminal-Bench v2.1
89%
SciCode
50%
Humanity’s Last Exam
37%
GPQA Diamond
89%
CritPt
20%
AA-Omniscience
63%
AA-LCR
68%

Similar models