LLM EXPLORER 59,420 MODELS INDEXED

QwQ 32B 8.0bpw H8 EXL2 by LoneStriker

By LoneStriker · 6 downloads
HF Repo ↗ Copy URL

QwQ 32B 8.0bpw H8 EXL2 is an open-source language model by LoneStriker. Features: 32b LLM, VRAM: 33.6GB, Context: 128K, License: apache-2.0, Quantized, LLM Explorer Score: 0.17.

  Arxiv:2309.00071   Arxiv:2407.10671   8-bit Base model:quantized:qwen/qwen...   Base model:qwen/qwen2.5-32b   Chat   Conversational   En   Endpoints compatible   Exl2   Quantized   Qwen2   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/LoneStriker/QwQ-32B-8.0bpw-h8-exl2 

QwQ 32B 8.0bpw H8 EXL2 Parameters and Internals

LLM NameQwQ 32B 8.0bpw H8 EXL2
Repository 🤗https://huggingface.co/LoneStriker/QwQ-32B-8.0bpw-h8-exl2 
Base Model(s)  Qwen/Qwen2.5-32B   Qwen/Qwen2.5-32B
Model Size32b
Required VRAM33.6 GB
Updated2026-07-15
MaintainerLoneStriker
Model Typeqwen2
Model Files  8.6 GB: 1-of-4   8.6 GB: 2-of-4   8.5 GB: 3-of-4   7.9 GB: 4-of-4
Supported Languagesen
Quantization Typeexl2
Model ArchitectureQwen2ForCausalLM
Licenseapache-2.0
Context Length131072
Model Max Length131072
Transformers Version4.43.1
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size152064
Torch Data Typebfloat16
Errorsreplace

Best Alternatives to QwQ 32B 8.0bpw H8 EXL2

Best Alternatives
Context / RAM
Downloads
Likes
...Logic Mix 3 32B Mlx OptiQ 5bit128K / 22.3 GB411
...pSeek R1 Distill Qwen 32B 4bit128K / 18.5 GB779551
Hydraulic Deepseek 16bit128K / 65.8 GB520
...k R1 Distill Qwen 32B Bnb 4bit128K / 19.2 GB271729
...till Qwen 32B Abliterated 4bit128K / 18.5 GB8887
QwQ 32B Unsloth Bnb 4bit128K / 22.5 GB17947
...ogito V1 Preview Qwen 32B 4bit128K / 18.5 GB435
...T1 Flash Lightest 32B Mlx 4Bit128K / 18.5 GB2318
QwQ 32B 4bit128K / 18.5 GB13238
QwQ 32B Bnb 4bit128K / 19.2 GB1574
Note: green Score (e.g. "73.2") means that the model is better than LoneStriker/QwQ-32B-8.0bpw-h8-exl2.