GLM5.2 vs DeepSeek in Writing
burny_tech · x · 2026-07-10
The cited post argues that while GLM5.2 excels at coding, its writing capabilities are significantly inferior to DeepSeek V4 Pro—not just by a small margin, but by a whole tier.
The original text uses the metaphor "a basement dweller at Tsinghua vs. a Peking University genius" to describe their stylistic differences, and recommends testing them via API with two tasks:
- Writing a book outline
- Writing a romance novel
The author believes these two tasks best highlight the difference in writing temperament between the two models.
More from Models
- Gary Marcus says LLM math skills are like knowing only a car’s engine size — GaryMarcus · 2026-07-22
- OpenAI’s Codex + GPT-5.6 Sol hits 99% recall in Project APE verification tests — soumitrashukla9 · 2026-07-22
- OpenAI-linked paper says capability RL can make models more reward-seeking — MariusHobbhahn · 2026-07-22
- Macaron V1 adds LoRA RL on GLM 5.2 and claims SOTA benchmark gains — Xianbao_QIAN · 2026-07-22
- OpenAI rolls out voice in GPT-Live, but the UI obscures search and reasoning — Graham_dePenros · 2026-07-22
- Moonshot’s Kimi K3 sets a new open-weights ECI record at 156 — scaling01 · 2026-07-22