LLM EXPLORER 56,604 MODELS INDEXED

GLM 4.5 Air 6bit by mlx-community

By mlx-community · 201 downloads
HF Repo ↗ Copy URL

GLM 4.5 Air 6bit is an open-source language model by mlx-community. Features: 106.9b LLM, VRAM: 86.7GB, Context: 128K, License: mit, Quantized, LLM Explorer Score: 0.2.

  6-bit   6bit Base model:quantized:zai-org/g...   Base model:zai-org/glm-4.5-air   Conversational   En   Glm4 moe   Mlx   Quantized   Region:us   Safetensors   Sharded   Tensorflow   Zh
Model Card on HF 🤗: https://huggingface.co/mlx-community/GLM-4.5-Air-6bit 

GLM 4.5 Air 6bit Parameters and Internals

LLM NameGLM 4.5 Air 6bit
Repository 🤗https://huggingface.co/mlx-community/GLM-4.5-Air-6bit 
Base Model(s)  zai-org/GLM-4.5-Air   zai-org/GLM-4.5-Air
Model Size106.9b
Required VRAM86.7 GB
Updated2026-07-16
Maintainermlx-community
Model Typeglm4_moe
Model Files  5.2 GB: 1-of-18   5.1 GB: 2-of-18   5.0 GB: 3-of-18   5.1 GB: 4-of-18   5.1 GB: 5-of-18   5.0 GB: 6-of-18   5.1 GB: 7-of-18   5.1 GB: 8-of-18   5.0 GB: 9-of-18   5.1 GB: 10-of-18   5.1 GB: 11-of-18   5.0 GB: 12-of-18   5.1 GB: 13-of-18   5.1 GB: 14-of-18   5.0 GB: 15-of-18   5.1 GB: 16-of-18   5.0 GB: 17-of-18   0.5 GB: 18-of-18
Supported Languagesen zh
Quantization Type6bit
Model ArchitectureGlm4MoeForCausalLM
Licensemit
Context Length131072
Model Max Length131072
Transformers Version4.54.0
Tokenizer ClassPreTrainedTokenizerFast
Padding Token<|endoftext|>
Vocabulary Size151552
Torch Data Typebfloat16

Best Alternatives to GLM 4.5 Air 6bit

Best Alternatives
Context / RAM
Downloads
Likes
INTELLECT 3 4bit128K / 60.4 GB143
GLM 4.5 Air 2bit128K / 33.4 GB1074
GLM 4.5 Air 8bit128K / 114.1 GB272611
GLM 4.5 Air 4bit128K / 60.4 GB126128
GLM 4.5 Air 3bit128K / 46.6 GB62132
GLM 4.5 Air 3bit DWQ V2128K / 46.6 GB824
GLM 4.5 Air 3bit DWQ128K / 53.3 GB504
GLM 4.5 Air 2bit DWQ128K / 40.2 GB282
INTELLECT 3128K / 194.3 GB6275216
GLM 4.5 Air Mxfp4128K / 56.6 GB582
Note: green Score (e.g. "73.2") means that the model is better than mlx-community/GLM-4.5-Air-6bit.