GLM 5.3 Criticized for Unnatural Language, Over-training Side Effects
askerlee · x · 2026-08-15
A user comments that GLM 5.3 speaks unnaturally, worse than 5.2, possibly due to over-training. Earlier discussions noted GLM 5.3 uses same base as 5.2 with significant post-training gains.
More from Models
- Meta quietly releases Muse Spark Open: 816B-param MoE with 42B active, on Hugging Face — victormustar · 2026-10-16
- Baseten serves Meta's Muse Spark Open on day 0 at $1.25/M input tokens — baseten · 2026-10-16
- Anthropic red team's first-ever model "abliteration" cut GLM-5.3 refusals from 90% to 3% for ~$4,400 — cephaloform · 2026-10-16
- d1-ultra tops Artificial Analysis Decision Index at ~$30 per 1M tasks — JosephJacks_ · 2026-10-16
- Developer to Google: loosen usage limits and merge your AI apps into one — dreamwieber · 2026-10-16
- Users frustrated as Gemini web app access doesn't carry over to coding tools — dreamwieber · 2026-10-16