Google Releases Gemini 4 Neon for Complex Software Engineering and Knowledge Work
Google has released Gemini 4 Neon, designed to rapidly complete end-to-end tasks across complex software engineering, knowledge work, and scientific workflows. The model supports multi-file debugging and refactoring, multi-step instruction following, a 1M-token context window, and four thinking levels, and recorded its lowest hallucination rate in internal Google Antigravity testing. In a demonstration, Gemini 4 Neon ran /swarm for 12 hours to reconstruct Alan Turing's 1936 research from archival photographs and build a working universal Turing machine.
Why read this
SelectedAI score 79
The materials highlight multi-file debugging, a 1M-token context window, and four thinking levels, making it easier to compare how the model handles complex engineering tasks.
Topics
Body
Meet Gemini 4 Neon — our newest model that brings frontier-level intelligence with fast end-to-end task completion across complex software engineering, knowledge work, and science workflows.
Neon brings frontier-level performance to your everyday work and is built to handle complex coding and problem-solving without the wait.
How, you might ask? ⬇️
— Smarter coding: Multi-file debugging and refactoring with targeted, surgical edits — Grounded answers: Excels at multi-step instruction following and builds on our ongoing progress to curb fabrication, recording our lowest fabrication rate yet in internal Google Antigravity testing — 1M-token context window: Synthesize massive projects, tables, and charts at once — Adjustable thinking: Balance and adjust speed and depth on the fly with 4 thinking levels
Watch Gemini 4 Neon run a 12-hour /swarm in @antigravity to reconstruct Alan Turing’s 1936 study from archival photos — complete with a running Universal Turing Machine, the blueprint for modern computing.
AI Trends