GLM-5.3-Flash Released: 320B Total Params, Cost-Efficiency on Pareto Frontier
ArtificialAnlys · x · 2026-08-27
Zai has released GLM-5.3-Flash, a smaller and more cost-efficient sibling to GLM-5.3. It features 320B total parameters with 18B active parameters and supports low, high, and max reasoning efforts.
In Artificial Analysis benchmarks, GLM-5.3-Flash scores 57 on the Intelligence Index and sits comfortably on the Intelligence vs. Cost per Task Pareto frontier at $0.09 per task. It achieves a GDPval-AA v2 score of 1770, matching the GLM-5.3 Max. The evaluation used 149M output tokens, approximately 11% fewer than the GLM-5.3 model.
Related event: Zhipu Open-Sources GLM-5.3-Flash: 320B MoE Multimodal at 1/10th the Cost(48 posts)→
More from Models
- Unredacted Anthropic filing exposes internal data-quality debate months before Opus 3 cutoff — zacharynado · 2026-10-15
- 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