Qwen3-Next-80B Thinking criticized for extreme verbosity and slow tool use
rebellioninmypants · reddit · 2026-08-21
A user reports that the Qwen3-Next-80B-A3B-Thinking model is excessively verbose during inference, frequently outputting filler words like "Alternatively" and "Wait". Even simple queries trigger long thinking delays (up to minutes), and its performance as a GitHub Copilot backend suffers from 2-3 minute hallucinations before actual file reading. The user contrasts this with the smoother experience of Qwen3.5-35B-A3B and provides their full llama-server configuration, asking if this is a design flaw or a configuration issue.
More from Models
- Decision models match LLMs on routing and safety — and are 7-8x faster — randal_olson · 2026-10-16
- Xiaomi MiMo V2.6 Pro takes #1 open-model spot on Agent Arena at $0.11/task — arena · 2026-10-16
- ChatGPT's 'best AI video tool' answer points to Runway integration, sparking ad doubts — tlakomy · 2026-10-16
- Microsoft's MAI finishes in-house 3T-parameter pretrained base models, post-training results coming — soumitrashukla9 · 2026-10-16
- Hands-on with Gemini 4 Argon: faster than Opus 5.5 but hallucinates more and costs more — prasenx · 2026-10-16
- Bindu Reddy urges Meta to open-source Wonder, claiming it beats DeepSeek — bindureddy · 2026-10-16