VERIFIEDOct 132026
Is GPT-6 Astra state of the art?
Yes.SOTA in co-SOTA in Reasoning and co-SOTA in Vision.
As of Oct 13, 2026, GPT-6 Astra from OpenAI. SOTA in co-SOTA in Reasoning and co-SOTA in Vision. GPT-6 Astra is co-SOTA in Reasoning with low confidence because primary sources disagree beyond the margin of error (with Claude Opus 5.5, Gemini 4 Argon).
ReasoningCo-SOTA#1 · 5dReasoning · paretoNot SOTA · #1#1 · 0dMathNot SOTA · #1#1 · 0dMath · paretoPareto#1 · 0dCodingNot SOTA · #5#5 · 0dCoding · paretoNot SOTA · #5#5 · 0dCoding › AgenticNear-SOTA#3 · 0dCoding › Agentic · paretoNear-SOTA#3 · 0dCoding › Web devNear-SOTA#2 · 0dCoding › Web dev · paretoNear-SOTA#2 · 0d
Why
The ruling, in plain English
GPT-6 Astra is co-SOTA in Reasoning with low confidence because primary sources disagree beyond the margin of error (with Claude Opus 5.5, Gemini 4 Argon).
Receipts
Evidence
| Source | Score | Date | Type | Receipt |
|---|---|---|---|---|
| Humanity's Last Exam gpt-6-astra_unknown | 0.5479999999999999 0.5–0.6 | 2026-09-03 | INDEP. | snap_f9d8103e52c2 |
| Terminal-Bench 4.0 GPT-6 Astra max · max | 58.18 55.4–61.0 | 2026-10-06 | INDEP. | snap_13678220dcf7 |
| LMArena WebDev gpt-6-astra-max · max | 1786.1141116447413 1775.2–1797.1 | 2026-10-08 | INDEP. | snap_6c327280fd30 |
| GDP.pdf gpt-6-astra_max · max | 0.342 | 2026-09-03 | INDEP. | snap_f9d8103e52c2 |
Reign history
Where it sits in the succession
Reasoning
Gemini 1.5 Pro 002
Gemini 2.5 Pro Exp 03.25
Gemini 2.5 Pro Preview 05.06
Gemini 2.5 Pro Preview 06.05
O3
O4 Mini
Gemini 2.5 Pro Preview 06.05
GPT-5
GPT-5 Mini
O3
Gemini 3 Pro Preview
GPT-5 Pro
Gemini 2.5 Pro
Gemini 3 Pro Preview
GPT-5 Pro
Claude Opus 4.1
Gemini 3 Pro Preview
GPT-5 Pro
Claude Opus 4.1
Claude Opus 4.6
Gemini 3 Pro Preview
GPT-5 Pro
Claude Opus 4.6
Gemini 3 Pro Preview
GLM 5
GPT-5 Pro
Claude Opus 4.6
Claude Sonnet 4.5
Gemini 3 Pro Preview
GPT-5 Pro
Claude Sonnet 4.5
Gemini 3.1 Pro Preview
Claude Sonnet 4.5
Gemini 3.1 Pro Preview
GPT-5.4 Pro
Gemini 3.1 Pro Preview
GPT-5.4 Mini
GPT-5.4 Pro
Gemini 3.1 Pro Preview
GPT-5.4 Mini
GPT-5.4 Pro
Muse Spark
Claude Opus 4.5
Gemini 3.1 Pro Preview
GPT-5.4 Pro
Muse Spark
Gemini 3.1 Pro Preview
GLM 5.1
GPT-5.4 Pro
Muse Spark
Gemini 3.1 Pro Preview
GPT-5.4 Pro
GPT-5.5
Muse Spark
Claude Sonnet 4.6
Gemini 3.1 Pro Preview
GPT-5.4
GPT-5.4 Pro
Muse Spark
Gemini 3.1 Pro Preview
Gemini 3 Pro
GPT-5.4 Pro
Muse Spark
Gemini 3.1 Pro Preview
GPT-5.4 Pro
Kimi K3
Muse Spark
Gemini 3.1 Pro Preview
GLM 5.3
GPT-5.4 Pro
Muse Spark
Claude Fable 5.1
Gemini 3.1 Pro Preview
Gemini 3.7 Flash
GPT-5.4 Pro
Muse Spark
Claude Fable 5.1
Gemini 3.1 Pro Preview
Gemini 3.7 Flash
Gemini 3.8 Flash
GPT-5.4 Pro
Muse Spark
Gemini 3.7 Flash
GPT-6 Astra
GPT-6 Astra
Muse Spark 1.1
Claude Fable 5
GPT-6 Astra
Claude Opus 5
GPT-6 Astra
Claude Opus 5.5
Gemini 4 Argon
GPT-6 Astra
Vision
Nova 2.0 Pro Preview
Claude Opus 4.7
Claude Sonnet 4.6
Gemini 3.1 Pro Preview
Claude Fable 5
GPT-5.6 Sol
GPT-5.6 Sol
GPT-6 Astra
Honest about the gaps
Where it is not the crown
| Category | This model | Why |
|---|---|---|
| Math | Not SOTA · #1 | Ranked #1 in Math. |
| Coding | Not SOTA · #5 | Ranked #5 in Coding. |
| Coding › Agentic | Near-SOTA | Close in Coding › Agentic: rank #3, 16.7 percentile points behind the leader. |
| Coding › Web dev | Near-SOTA | Close in Coding › Web dev: rank #2, 12.5 percentile points behind the leader. |