LLM EXPLORER 61,561 MODELS INDEXED

DeepSeek R1 Qwen3 0528 8B 4bit AWQ by mlx-community

By mlx-community · 80 downloads
HF Repo ↗ Copy URL

DeepSeek R1 Qwen3 0528 8B 4bit AWQ is an open-source language model by mlx-community. Features: 8b LLM, VRAM: 4.7GB, Context: 128K, License: mit, Quantized, LLM Explorer Score: 0.19.

  4bit   Awq Base model:deepseek-ai/deepsee... Base model:finetune:deepseek-a...   Conversational   Mlx   Quantized   Qwen3   Region:us   Safetensors
Model Card on HF 🤗: https://huggingface.co/mlx-community/DeepSeek-R1-Qwen3-0528-8B-4bit-AWQ 

DeepSeek R1 Qwen3 0528 8B 4bit AWQ Parameters and Internals

LLM NameDeepSeek R1 Qwen3 0528 8B 4bit AWQ
Repository 🤗https://huggingface.co/mlx-community/DeepSeek-R1-Qwen3-0528-8B-4bit-AWQ 
Base Model(s)  DeepSeek R1 0528 Qwen3 8B   deepseek-ai/DeepSeek-R1-0528-Qwen3-8B
Model Size8b
Required VRAM4.7 GB
Updated2026-08-07
Maintainermlx-community
Model Typeqwen3
Model Files  4.7 GB
AWQ QuantizationYes
Quantization Typeawq|4bit
Model ArchitectureQwen3ForCausalLM
Licensemit
Context Length131072
Model Max Length131072
Transformers Version4.51.0
Tokenizer ClassLlamaTokenizerFast
Padding Token<|vision_pad|>
Vocabulary Size151936
Torch Data Typebfloat16

Best Alternatives to DeepSeek R1 Qwen3 0528 8B 4bit AWQ

Best Alternatives
Context / RAM
Downloads
Likes
...Seek R1 0528 Qwen3 8B AWQ 4bit128K / 6 GB5874
Qwen3 8B AWQ40K / 6 GB103068052
Gen Sql 8B Thinking AWQ40K / 7.5 GB50
Qwen3 8B 4bit AWQ40K / 4.7 GB1861
BehChat SFTv3 Ckpt1128K / 16.4 GB60
BehChat SFTv2 Mixed Ckpt 6128K / 16.4 GB110
...0528 Qwen3 8B Unsloth Bnb 4bit128K / 7.5 GB445713
DeepSeek R1 0528 Qwen3 8B 4bit128K / 4.6 GB17816
...Seek R1 0528 Qwen3 8B Bnb 4bit128K / 6.1 GB9239
Qwen3 8B LQA 14e Full128K / 16.4 GB60
Note: green Score (e.g. "73.2") means that the model is better than mlx-community/DeepSeek-R1-Qwen3-0528-8B-4bit-AWQ.