banteg benchmarks unreleased Gemini 4 Hydrogen and Neon, mocks Google's model lineup
banteg · x · 2026-11-19
banteg says he is evaluating two unreleased models, 'Gemini 4 Hydrogen' and 'Gemini 4 Neon', without a livestream this time; results and replays will follow. Screenshots show the two models on a pixel-art puzzle benchmark: Neon at 20/66 rooms, Hydrogen at 16/66, same 18:39 active time but Hydrogen burning more turns and inputs.
He also asks for help decoding Google's ever-multiplying Gemini model positioning.
More from Models
- GPT-6.2 scores worse than predecessors across the board on Company Agent Bench — philhchen · 2026-11-19
- Qwen4-Max launches with Day-0 support in Qoder coding IDE — usamawahabkhan · 2026-11-19
- RoboLab-120 model runs autonomous lab experiments, beats GPT Astra by ~10 points — Scobleizer · 2026-11-19
- Google's Gemini 4 Neon goes GA in GitHub Copilot, tuned for agentic coding — intellectronica · 2026-11-19
- Ex-bootcamp teacher: frontier models code like talented bootcamp students — dansitu · 2026-11-19
- Gemini 4 Neon wows with Triceratops skeleton rendering — fofrAI · 2026-11-19