Gemini re-renders images as text when vision is removed, dropping only 6% WSR
zacharynado · x · 2026-11-12
Experiments by bcaine show vision matters: removing the viewimage tool drops WSR 6% and increases token consumption 1.85x. But zacharynado observed something wild—when Gemini's vision input is taken away, it simply re-renders the image in text and feeds it to itself as text input, showing shocking persistence in solving tasks.
More from Models
- Mystery Google model 'Gemini Decision 1' spotted on model tracker — imjustnewatai · 2026-11-12
- AI in 24 hours: Gemini 4 Carbon, Claude Whiteboards, Qwen4-Max preview, Cursor multiplayer agents — minchoi · 2026-11-12
- DeepSeek V4.1 Flash runs at 20 tok/s on a single RTX 3090 with 64GB RAM — casper_hansen_ · 2026-11-12
- Googler casually mentions 'our first Gemini 4 models' — strongest public hint yet of Gemini 4 — Div_pradeep · 2026-11-12
- Robot build test: GPT-6 Astra stabilizes two-board structure, Claude Fable 5.2 falls short — qinzytech · 2026-11-12
- Zvi dissects Anthropic's latest system card, disputing the autonomy threshold call — Don't Worry About the Vase (Zvi) · 2026-11-12