Geo News
Community curated by people like you
LatestAICryptoHealthWorld AffairsUS Politics
Google launches Gemini 4 Neon, its fourth new model in a month
00

Google launches Gemini 4 Neon, its fourth new model in a month

Nov 18, 2026

Google launched Gemini 4 Neon on November 18, 2026, marking its fourth new model release in a single month. Positioned between Gemini 4 Hydrogen and Carbon, Neon is available via Google's API at $3 per million input tokens and $15 per million output tokens. Neon features a one-million-token context window supporting multimodal inputs. On benchmarks, Neon scored 61 on the Artificial Analysis Intelligence Index and 70.9% on the Vals Index, outperforming competing proprietary models while streaming output at 110 tokens per second.

Launch details and model releases

  • ▪Google released Gemini 4 Neon on November 18, 2026, its fourth new Gemini 4 model release within a month, across the paid Gemini API in Google AI Studio and Android Studio, the Gemini app, and Google Antigravity platform
  • ▪Prior to releasing Gemini 4 Neon on November 18, 2026, Google launched Gemini 4 Argon seven weeks earlier, followed by Gemini 4 Carbon on October 28, 2026, and Gemini 4 Hydrogen

Model capabilities and platform integration

  • ▪Gemini 4 Neon features a context window of approximately one million tokens and handles multimodal inputs including text, images, video, audio, and PDFs
  • ▪Google Pro tier subscribers can select Gemini 4 Neon in the Gemini app, while access to Gemini 4 Carbon requires an Ultra subscription
  • ▪Google positions Gemini 4 Hydrogen as its high-volume workhorse model, Gemini 4 Neon for recurring complex tasks, and Gemini 4 Carbon for long-horizon projects

API pricing structure and comparisons

  • ▪Gemini 4 Neon is priced 25% below Anthropic's flagship Claude Opus 5.5 standard API rates of $4 per million input tokens and $20 per million output tokens
  • ▪Gemini 4 Neon headline API rates are 50% higher than standard short-context rates for Anthropic's Claude Sonnet 5.5 and OpenAI's GPT-6.2 Sol, which both charge $2 per million input and $10 per million output tokens
  • ▪Starting December 1, 2026, Gemini 4 Neon's input and output token prices will be 25% below Gemini 4 Carbon's scheduled non-introductory rates of $4 per million input and $20 per million output tokens
  • ▪Gemini 4 Neon API token pricing is set at $3 per million input tokens and $15 per million output tokens

Gemini 4 Neon speed metrics

  • ▪Benchmark tracker Priors recorded Gemini 4 Neon streaming approximately 110 tokens per second at its extra-high reasoning setting, compared to 59 tokens per second for Gemini 4 Carbon on high effort
  • ▪Benchmark tracker Priors recorded a delay of approximately 57 seconds to the first token for Gemini 4 Neon at its extra-high reasoning setting

3 sources

VentureBeat
Google launches Gemini 4 Neon with Opus 5.5-level coding performance at 25% lower API token prices
View source article
The Verge
Google released another Gemini 4 model.
View source article
The Decoder
Google ships Gemini 4 Neon, its fourth new Gemini model in a month
View source article

Featured stories

View more in GPT-6.1 Sol

OpenAI launches GPT-6.2 Astra and Sol models to mixed reception

Nov 17, 2026 · 4 sources

Microsoft launches MAI Code 2 and MAI Cyber 2 Flash AI models

Nov 17, 2026 · 3 sources

Alibaba unveils Qwen4-Max flagship AI model

Nov 16, 2026 · 2 sources

OpenAI and AMD cut chip design time from a year to under five months using AI

Nov 18, 2026 · 2 sources

Story comments

Loading comments…

Related entities

FrontierCode 2.0Claude Opus 4.8

Related Projects

GoogleAnthropicArtificial AnalysisMini SWE-agentCognition

Topics

GPT-6.1 SolFrontier modelsClaude Sonnet 5.5AI benchmarksGoogle AI StudioAI ModelsGemini 4 ArgonOpenAIClaude Opus 5.5Gemini 4 Carbon

Featured stories

View more in GPT-6.1 Sol

OpenAI launches GPT-6.2 Astra and Sol models to mixed reception

Nov 17, 2026 · 4 sources

Microsoft launches MAI Code 2 and MAI Cyber 2 Flash AI models

Nov 17, 2026 · 3 sources

Alibaba unveils Qwen4-Max flagship AI model

Nov 16, 2026 · 2 sources

OpenAI and AMD cut chip design time from a year to under five months using AI

Nov 18, 2026 · 2 sources