Gemini Users Lament Flash 3.6 Replacement by Auto: Dumber Replies and Bugs
JD__052160 · reddit · 2026-10-18
A longtime Gemini user details a collapse in experience after Flash 3.6 was replaced by Auto/Flash-lite 3.5 on the 9th: markedly worse answers, the model questioning simple prompts, broken personalization, ignored instructions, plus sync bugs across devices, irrelevant follow-up questions and random off-topic endings — none fixed by reinstalling.
The author says Flash was the reason they left ChatGPT, and no Auto tier (low/medium/high) recovers the old quality.
More from Models
- Qwen3.8 27B benchmarks on NVIDIA RTX Spark: 19.3 tok/s decode, 100% accuracy — draginol · 2026-10-19
- Allie Miller: Claude leads on models, Codex on features — she uses both — alliekmiller · 2026-10-19
- Gemma 4 12b on NVIDIA RTX Spark laptop: 31.6 tok/s decode, 100% accuracy — draginol · 2026-10-19
- Same prompt, three designs: Claude vs ChatGPT vs Figma Make UI face-off — Tegadesigns · 2026-10-19
- Mac Studio M5 Ultra beats dual DGX Sparks in local LLM decode speeds — SumitGup · 2026-10-19
- Three mystery OpenAI models (Aurora red/green/alpha) spotted; likely image models, GPT Image refresh imminent — kimmonismus · 2026-10-19