Zhipu's GLM 5.3 Gains on Same Base Model, Highlighting Post-Training Power
oran_ge · x · 2026-08-15
The user praises Zhipu as the 'god of post-training,' noting that GLM 5.3 shows significant improvements over 5.2 while sharing the same 756B base model, indicating vast room for post-training optimization. Citing DeepSeek V4 Flash's 304B model outperforming GLM 5.2, the user speculates that Luna is around 500B and asserts OpenAI's mastery of post-training as well.
More from Models
- Google says teams across the company are using Gemini 4 Argon on its hardest engineering problems — CurieuxExplorer · 2026-10-15
- Perplexity's new embedding models: index with a 9B model, search with 0.6B — antoine_chaffin · 2026-10-15
- 797GB of n-gram counts in CPU RAM let 1B models match 57% more compute — PangWeiKoh · 2026-10-15
- Leak: Moonshot's Kimi flagship 'x4' done with pre-training, K3.1 in internal testing — lyraxana · 2026-10-15
- Unsloth ships GGUF quants for Nemotron 3.5: Bolt-74B runs in 24GB RAM — danielhanchen · 2026-10-15
- NVIDIA ships two open Nemotron 3.5 Bolt models: a 74B-A9B MoE and a mobile-friendly E4B — HankYeomans · 2026-10-15