Dev reports: Qwen 3.8 27B only works with reasoning_level set to low
andrejusb · x · 2026-08-20
A developer reported that the Qwen 3.8 27B model is only usable when reasoninglevel is set to low. Setting it to medium or higher causes the model to "think" for too long, making it impractical.
More from Models
- Superwhisper launches S1-mini, a 0.6B on-device model for transcripts — Scobleizer · 2026-08-20
- Agent Arena: Top Agent Models Differ Over 5x in Cost per Task — arena · 2026-08-20
- Agent Arena: Kimi K3 offers best value, Claude Opus 5 is priciest — arena · 2026-08-20
- Model 'Thinking' is Just Post-Training with Thought Sections — majidmanzarpour · 2026-08-20
- Overly Colorful Verbs: A Tell for AI Writing — dbasch · 2026-08-20
- Why Qwen3.8 27B costs more than DSv4 Flash? KV cache and quantization explained — ihatebeinganonymous · 2026-08-20