LLM EXPLORER 62,404 MODELS INDEXED

GLM 5.3 Flash 4bit by mlx-community

By mlx-community · 364 downloads
HF Repo ↗ Copy URL

GLM 5.3 Flash 4bit is an open-source language model by mlx-community. Features: 347.9b LLM, VRAM: 150.7GB, License: mit, Quantized, LLM Explorer Score: 0.47, ELO: 1473.

  4-bit   4bit Base model:quantized:zai-org/g... Base model:zai-org/glm-5.3-fla...   Conversational   Glm5 next   Image-text-to-text   Mlx   Quantized   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/mlx-community/GLM-5.3-Flash-4bit 

GLM 5.3 Flash 4bit Benchmarks

LMSys ELO 1473 vs 1272 (so35) 15.8%
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

GLM 5.3 Flash 4bit Parameters and Internals

LLM NameGLM 5.3 Flash 4bit
Repository 🤗https://huggingface.co/mlx-community/GLM-5.3-Flash-4bit 
Base Model(s)  zai-org/GLM-5.3-Flash   zai-org/GLM-5.3-Flash
Model Size347.9b
Required VRAM150.7 GB
Updated2026-09-28
Maintainermlx-community
Model Typeglm5_next
Model Files  4.3 GB: 1-of-62   3.4 GB: 2-of-62   3.3 GB: 3-of-62   3.4 GB: 4-of-62   3.3 GB: 5-of-62   3.2 GB: 6-of-62   3.4 GB: 7-of-62   3.4 GB: 8-of-62   3.2 GB: 9-of-62   3.4 GB: 10-of-62   3.3 GB: 11-of-62   3.4 GB: 12-of-62   3.2 GB: 13-of-62   3.4 GB: 14-of-62   3.4 GB: 15-of-62   3.2 GB: 16-of-62   3.4 GB: 17-of-62   3.4 GB: 18-of-62   3.4 GB: 19-of-62   3.2 GB: 20-of-62   3.4 GB: 21-of-62   3.3 GB: 22-of-62   3.2 GB: 23-of-62   3.4 GB: 24-of-62   3.4 GB: 25-of-62   3.4 GB: 26-of-62   3.2 GB: 27-of-62   3.3 GB: 28-of-62   3.4 GB: 29-of-62   3.2 GB: 30-of-62   3.4 GB: 31-of-62   3.3 GB: 32-of-62   3.4 GB: 33-of-62   3.2 GB: 34-of-62   3.3 GB: 35-of-62   3.4 GB: 36-of-62   3.2 GB: 37-of-62   3.4 GB: 38-of-62   3.4 GB: 39-of-62   3.3 GB: 40-of-62   3.2 GB: 41-of-62   3.4 GB: 42-of-62   3.4 GB: 43-of-62   3.2 GB: 44-of-62   3.4 GB: 45-of-62
Quantization Type4bit
Model ArchitectureGlm5NextForConditionalGeneration
Licensemit
Transformers Version5.16.0