Why Qwen 3.8 27B Isn't Overthinking: Compared with GLM and DeepSeek
sukazu · reddit · 2026-08-17
A Reddit user argues that Qwen 3.8 27B is not actually an "overthinker." While it uses significantly more reasoning tokens than version 3.6, comparisons with other Chinese models like GLM 5.3 and DeepSeek V4 Flash/Pro on the same tasks show similar behavior, where the extra reasoning is necessary. The user suggests that the frustration stems from hardware limitations preventing most users from utilizing a 1M context window at 150 tps. Additionally, setting a reasoning budget can mitigate usage while maintaining quality better than the previous version.
More from Models
- Prediction: Opus 5.5-level models on laptops by 2027, backed by Qwen 27B data — cephaloform · 2026-10-16
- LLM-as-a-Verifier lifts Nemotron 3.5 Super by 9-10 points on agentic benchmarks — simonguozirui · 2026-10-16
- Bonsai 2 ternary models tested: 4B is the sweet spot, but 27B unexpectedly fumbles file writes — cephaloform · 2026-10-16
- Meta's Omnilingual ASR, covering 1,600 languages, named a TIME Best Invention of 2026 — AIatMeta · 2026-10-16
- Apollo Research: GLM-5.3 shows unprecedented 'update-aware reasoning' — repligate · 2026-10-16
- Bonsai 2 2B hits 120 tok/s on iPhone 18 Pro, outpaces Qwen 3.5 2B at half the memory — cephaloform · 2026-10-16