LLMrank

The AI models we measure with

Gemini 3.8 Flash

Google

Google's Gemini models answer across Google's own surfaces and apps. Gemini 3.8 Flash is the current fast Gemini we measure - we read how that ecosystem describes and recommends your company.

Model details

Model ID
gemini-3.8-flash
Context window
1M tokens
Knowledge cutoff
Mar 2026

Knowledge cutoff vs training data cutoff

Knowledge cutoff and training data cutoff are different dates: the knowledge cutoff is roughly how current a model's world knowledge is, while the training data cutoff is when its training data ends. Both shape what a model knows about you.

What we use this model for

We ask this model directly about your company to see how it describes you and which names it brings up in your space. It is one of several frontier models in our measurement set, so your visibility is read across the assistants people actually use rather than from a single model.

Recent updates for this model

  1. today

    Google DeepMind released Gemini 4 Carbon, a high-intelligence model specializing in cybersecurity and secure code writing that outperforms Claude Opus 5.5 and GPT-6 Astra in benchmarks and reached the top of the Artificial Analysis Intelligence Index v5. Google also launched the Suncatcher prototype space data center, which utilizes Tensor Processing Units and Gemma 4 to perform inference workloads in orbit. Google updated Gemini to enable direct generation and editing of PDFs and Docs within the chat interface and integrated Gemini into Chrome autofill to extract order and flight details from Gmail. Alphabet reports that AI search usage is currently growing faster than corresponding ad revenue.

    • Google DeepMind released Gemini 4 Carbon, a cybersecurity-specialized model that reached the top of the Artificial Analysis Intelligence Index v5.
    • Google launched the Suncatcher prototype space data center using Tensor Processing Units and Gemma 4 to perform orbital inference workloads.
    • Google updated Gemini to allow users to generate and edit PDFs and Docs directly within the chat interface.
    • Google updated Chrome to enable autofill to pull order numbers and flight details from Gmail using Gemini.
    • Gemini 4 Carbon outperformed Claude Opus 5.5 and GPT-6 Astra in performance benchmarks.

    Sources: theverge.com, techcrunch.com, ComputerSweden@edt.computersweden.se, aigenart@mail.beehiiv.com

  2. 1 days ago

    Google released Gemini 4 Argon, which demonstrates specific strengths in organizing complex topics. Gemini 4 Carbon performance metrics were also reported. Ben Collins launched a 'Gemini for Builders' course focused on specification-driven development for app creation.

    • Google released Gemini 4 Argon, which specializes in organizing complex topics.
    • Google reported performance data for Gemini 4 Carbon.
    • Ben Collins launched the 'Gemini for Builders' course to teach specification-driven development for AI-generated apps.

    Sources: hello@every.to, hello@8020ai.co, ben@benlcollins.com

  3. 2 days ago

    Google released Gemini 4 Carbon, a frontier model specializing in cybersecurity, coding, and agentic tasks. The company also launched Gemini 4 Argon and integrated Gemini with Personal Intelligence into Chrome's autofill feature to retrieve form data from Gmail.

    • Google integrated Gemini with Personal Intelligence into Chrome's autofill to suggest form data sourced from Gmail.
    • Google released Gemini 4 Carbon, a frontier model designed for cybersecurity, coding, and agentic tasks.
    • Google launched Gemini 4 Argon, coinciding with the release of the Pixel 11.

    Sources: theverge.com, techcrunch.com, superhuman@mail.joinsuperhuman.ai

View the provider's model card Visit the provider

See how these models describe your company.
Run an analysis