Pyyan / Compare / Gemini 4 Neon vs GPT-6 Astra

Gemini 4 Neon vs GPT-6 Astra

2 of 5

Language Models · verified 19 Nov 2026

Gemini 4 NeonGooglecurrent
GPT-6 AstraOpenAIcurrent
3 slots left
SpecificationGemini 4 NeonGPT-6 Astra
SummaryGoogle's middle Gemini 4, between Carbon and Hydrogen, at $3 and $15.Saturates three of the benchmarks it reports, and ships disabled for enterprise.
Context1M1.05M
Input ($/Mtok)$3$10
Output ($/Mtok)$15$50
LicenceProprietaryProprietary
Max output64K default, 1M with long decode128K
ParametersNot publishedNot published
Released18 Nov 20263 Sep 2026
Model IDgemini-4-neongpt-6-astra
CategoryLanguage ModelsLanguage Models
OfficialGoogle ↗—

Highlighted rows are where these differ.

Gemini 4 Neon

  • Released on 18 November 2026 in the Gemini API, Antigravity and the Gemini app for Pro and Ultra subscribers
  • $3 in, $0.15 cached and $15 out per million tokens; cheaper per token than Carbon once Carbon's introductory price ends on 30 November
  • Google says it scores 61 on the Artificial Analysis Intelligence Index at xhigh, third overall, and ranks second on the Vals Index
  • Four thinking levels; Google says it can direct parallel Hydrogen subagents in Antigravity
  • A cyber variant is limited to the Fairwind Program

Best for complex coding, analysis and agent work that needs to finish faster than Carbon.

Full spec sheet →

GPT-6 Astra

  • 97.6% on FrontierMath Tier 4, 99.9% on ARC-AGI-3 and 100% on ExploitBench
  • Above 272K input tokens the whole request reprices: input and cache at 2x, output at 1.5x
  • Rolls out to OpenAI's Daybreak cyber programme first, and arrives switched off for enterprise

Best for hard reasoning and computer use.

Full spec sheet →