LLM EXPLORER 59,962 MODELS INDEXED

GLM 5.3 Flash MLX Q9 by inferencerlabs

By inferencerlabs · 1093 downloads
HF Repo ↗ Copy URL

GLM 5.3 Flash MLX Q9 is an open-source language model by inferencerlabs. Features: LLM, VRAM: 352.2GB, Quantized, LLM Explorer Score: 0.45, ELO: 1259.

Base model:quantized:zai-org/g... Base model:zai-org/glm-5.3-fla...   Conversational   En   Glm5 next   Image-text-to-text   Mlx   Q9   Quantized   Region:us   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/inferencerlabs/GLM-5.3-Flash-MLX-Q9 

GLM 5.3 Flash MLX Q9 Benchmarks

LMSys ELO 1259 vs 1272 (so35) -1%
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 MLX Q9 Parameters and Internals

LLM NameGLM 5.3 Flash MLX Q9
Repository 🤗https://huggingface.co/inferencerlabs/GLM-5.3-Flash-MLX-Q9 
Base Model(s)  zai-org/GLM-5.3-Flash-BF16   zai-org/GLM-5.3-Flash-BF16
Required VRAM352.2 GB
Updated2026-08-30
Maintainerinferencerlabs
Model Typeglm5_next
Model Files  10.2 GB: 1-of-42   8.3 GB: 2-of-42   8.3 GB: 3-of-42   8.3 GB: 4-of-42   8.3 GB: 5-of-42   8.3 GB: 6-of-42   8.3 GB: 7-of-42   8.3 GB: 8-of-42   8.3 GB: 9-of-42   8.3 GB: 10-of-42   8.3 GB: 11-of-42   8.3 GB: 12-of-42   8.3 GB: 13-of-42   8.3 GB: 14-of-42   8.3 GB: 15-of-42   8.3 GB: 16-of-42   8.3 GB: 17-of-42   8.3 GB: 18-of-42   8.3 GB: 19-of-42   8.3 GB: 20-of-42   8.3 GB: 21-of-42   8.3 GB: 22-of-42   8.3 GB: 23-of-42   8.3 GB: 24-of-42   8.3 GB: 25-of-42   8.3 GB: 26-of-42   8.3 GB: 27-of-42   8.3 GB: 28-of-42   8.3 GB: 29-of-42   8.3 GB: 30-of-42   8.3 GB: 31-of-42   8.3 GB: 32-of-42   8.3 GB: 33-of-42   8.3 GB: 34-of-42   8.3 GB: 35-of-42   8.3 GB: 36-of-42   8.3 GB: 37-of-42   8.3 GB: 38-of-42   8.3 GB: 39-of-42   8.3 GB: 40-of-42   8.3 GB: 41-of-42   10.0 GB: 42-of-42
Supported Languagesen
Quantization Typeq9
Model ArchitectureGlm5NextForConditionalGeneration
Transformers Version5.16.0