GLM 4.5 Air 3bit DWQ is an open-source language model by mlx-community. Features: 106.9b LLM, VRAM: 53.3GB, Context: 128K, License: mit, Quantized, LLM Explorer Score: 0.22.
| LLM Name | GLM 4.5 Air 3bit DWQ |
| Repository ๐ค | https://huggingface.co/mlx-community/GLM-4.5-Air-3bit-DWQ |
| Base Model(s) | |
| Model Size | 106.9b |
| Required VRAM | 53.3 GB |
| Updated | 2025-10-28 |
| Maintainer | mlx-community |
| Model Type | glm4_moe |
| Model Files | |
| Supported Languages | en zh |
| Quantization Type | 3bit |
| Model Architecture | Glm4MoeForCausalLM |
| License | mit |
| Context Length | 131072 |
| Model Max Length | 131072 |
| Transformers Version | 4.54.0 |
| Tokenizer Class | PreTrainedTokenizerFast |
| Padding Token | <|endoftext|> |
| Vocabulary Size | 151552 |
| Torch Data Type | bfloat16 |
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| GLM 4.5 Air 4bit | 128K / 60.4 GB | 1055 | 24 |
| INTELLECT 3 4bit | 128K / 60.4 GB | 37 | 3 |
| GLM 4.5 Air 3bit | 128K / 46.6 GB | 353 | 31 |
| GLM 4.5 Air 8bit | 128K / 114.1 GB | 401 | 6 |
| GLM 4.5 Air 2bit | 128K / 33.4 GB | 256 | 4 |
| GLM 4.5 Air 2bit DWQ | 128K / 40.2 GB | 229 | 2 |
| GLM 4.5 Air 3bit DWQ V2 | 128K / 46.6 GB | 59 | 4 |
| GLM 4.5 Air 6bit | 128K / 86.7 GB | 111 | 2 |
| GLM 4.5 Air Mxfp4 | 128K / 56.6 GB | 43 | 2 |
๐ Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! ๐