Google details Gemini 4 Neon: 1M context, 4 thinking levels, lowest fabrication rate yet
GoogleAI · x · 2026-11-19
Google officially introduced Gemini 4 Neon, pairing frontier-level intelligence with fast end-to-end task completion for complex coding and problem-solving. Key points:
- Smarter coding: multi-file debugging and refactoring with targeted, surgical edits
- Grounded answers: excels at multi-step instruction following; records Google's lowest fabrication rate yet in internal Antigravity testing
- 1M-token context window: synthesize massive projects, tables, and charts at once
- Adjustable thinking: 4 thinking levels to balance speed and depth on the fly
In a demo, Neon ran a 12-hour /swarm in Antigravity to reconstruct Alan Turing's 1936 study from archival photos — complete with a running Universal Turing Machine.
Related event: Google Launches Gemini 4 Neon for Frontier-Speed Task Completion(25 posts)→
More from coding & agent
- Mythical Creature: A Design Studio That Designs Directly in Your Codebase, No Figma — round · 2026-11-19
- a16z Leads Series B for Linq, Betting on Communication Infrastructure for AI Agents — Scobleizer · 2026-11-19
- Gemini 4 Neon hits 59 on Coding Agent Index, matching Claude flagships at half the cost — ArtificialAnlys · 2026-11-19
- Gemini 4 Neon matches Claude Opus 5.5 on coding agents at roughly half the cost — ArtificialAnlys · 2026-11-19
- 100% local 5-person video character swap: H3MD 0.5 keyframe workflow — ReferenceFluffy6159 · 2026-11-19
- One prompt turns Claude Code into an autonomous web agent via Tilion + Fable 5.2 — Arindam_1729 · 2026-11-19