Reddit: Qwen3.8-Flash-Next preview is so strong that Qwen4 may make larger models unnecessary
spaceman_ · reddit · 2026-10-26
A Reddit thread on Qwen4 expectations: the poster doesn't try to predict model sizes, but argues that Qwen3.8-Flash-Next — a preview model meant to showcase the new architecture — is massively impressive for its size, suggesting Qwen4 could remove the need for larger models entirely. They ask whether Qwen4 will exceed the current QFN by a lot, or whether the current model is already near the optimum achievable at this size today.
More from Models
- Running a 125B model on 8GB VRAM with Strata: 900 t/s prompt processing, 30 t/s generation — BleynSpecnaz · 2026-10-26
- More OG Hugging Face profiles to follow: local AI, diffusion, kernels and transformer.js — ariG23498 · 2026-10-26
- 9 Hugging Face profiles worth following, from TRL agents to vLLM kernels — ariG23498 · 2026-10-26
- Terence Tao says OpenAI went 'code red' to beat Anthropic on Navier-Stokes, pausing $200 Pro signups — Hesamation · 2026-10-26
- LoopUno: 52-Minute LoRA Training Makes 3B LLM Draft 2 Tokens per Step Losslessly — LucaAmb · 2026-10-26
- OpenAI enterprise price cuts spark debate as users cite quota frustration and churn — teortaxesTex · 2026-10-26