DataLearner logo

GPT-6 AstravsGPT-5.5

Across 8 shared benchmarks, GPT-6 Astra leads overall: GPT-6 Astra wins 8, GPT-5.5 wins 0, with 0 ties and an average score difference of +17.40.

OpenAI
GPT-6 Astra

OpenAI · 2026-09-03 · Reasoning model

OpenAI
GPT-5.5

OpenAI · 2026-04-23 · Reasoning model

GPT-6 Astra8 wins(100%)(0%)0 winsGPT-5.5

Benchmark scores

Grouped by capability, sorted by largest gap within each. 8 shared benchmarks.

General Knowledge

GPT-6 Astra 4/4
BenchmarkGPT-6 AstraGPT-5.5Diff
ARC-AGI-362.701 / 16Max (No Tools)012 / 16Thinking High (No Tools)+62.70
ARC-AGI-2951 / 85Max (No Tools)8513 / 85Thinking High (No Tools)+10
HLE57.2012 / 190Max (With Tools)52.2027 / 190Thinking High (With Tools)+5
ARC-AGI-198.501 / 91Max (No Tools)9516 / 91Extra-High (No Tools)+3.50

AI Agent - Information Search

GPT-6 Astra 1/1
BenchmarkGPT-6 AstraGPT-5.5Diff
BrowseComp91.501 / 56Max (With Tools)84.409 / 56Thinking High (With Tools + Internet)+7.10

Coding and Software Engineer

GPT-6 Astra 1/1
BenchmarkGPT-6 AstraGPT-5.5Diff
DeepSWE74.102 / 35Max (With Tools)6711 / 35Extra-High (With Tools)+7.10

General Evaluation

GPT-6 Astra 1/1
BenchmarkGPT-6 AstraGPT-5.5Diff
GPQA Diamond961 / 271Max (No Tools)77.27162 / 271Normal (No Tools)+18.73

Math and Reasoning

GPT-6 Astra 1/1
BenchmarkGPT-6 AstraGPT-5.5Diff
FrontierMath Tier 4 v297.601 / 41Max (No Tools)72.506 / 41Extra-High (No Tools)+25.10

Specs

FieldGPT-6 AstraGPT-5.5
PublisherOpenAIOpenAI
Release date2026-09-032026-04-23
Model typeReasoning modelReasoning model
ArchitectureDenseDense
ParametersNot availableNot available
Context length1.05M1000K
Max output128K128K

API pricing

Prices use DataLearner records when available; missing fields are not inferred.

ItemGPT-6 AstraGPT-5.5
Text input$10 / 1M tokens$0.5 / 1M tokens
Text output$50 / 1M tokens$30 / 1M tokens
Cache read$1 / 1M tokens$0.5 / 1M tokens
Cache write$12.5 / 1M tokens$6.25 / 1M tokens

Summary

  • GPT-6 Astraleads in:General Knowledge (4/4), AI Agent - Information Search (1/1), Coding and Software Engineer (1/1), General Evaluation (1/1), Math and Reasoning (1/1)

On average across the 8 shared benchmarks, GPT-6 Astra scores 17.40 higher.

Largest single-benchmark gap: ARC-AGI-3 — GPT-6 Astra 62.70 vs GPT-5.5 0 (+62.70).

Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.