Try Gemini 4 Neon in the Workbench
Run this model interactively, tune parameters, and compare outputs.
gemini-4-neon
gemini-4-neon is a Multimodal LLM. It is Googleβs Gemini 4 model for fast end-to-end task completion across complex software engineering, knowledge work, and science workflows. Google highlights clearer writing, precise handling of uploaded spreadsheets and PDFs, and research that draws on current web sources.
Some other noteworthy features of gemini-4-neon include four thinking levels (low, medium, high, and xhigh, with high as the default), structured output, function calling, code execution, search grounding, URL context, file search, and computer use. It accepts text, images, audio, video, and PDFs and outputs text. Long decode extends a single response past 64K tokens, up to the full 1M token context shared with the input, for whole reports, codebases, and book-length drafts. It sits between Gemini 4 Hydrogen and Gemini 4 Carbon, and suits coding agents, document analysis, and multi-step research where both depth and turnaround time matter.
Example request
Fetch model details
The models endpoint returns the full model object, including itsjson_request_schema.