What are the hardware requirements for Huihui Qwen3.8 27B Abliterated AWQ MTP?

Huihui Qwen3.8 27B Abliterated AWQ MTP requires approximately 19.1 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Huihui Qwen3.8 27B Abliterated AWQ MTP and how large is it?

Huihui Qwen3.8 27B Abliterated AWQ MTP is developed by shawnw3i, a model with 27b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Huihui Qwen3.8 27B Abliterated AWQ MTP?

Huihui Qwen3.8 27B Abliterated AWQ MTP is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Huihui Qwen3.8 27B Abliterated AWQ MTP by shawnw3i — VRAM 19.1GB

Name: Huihui Qwen3.8 27B Abliterated AWQ MTP
Author: shawnw3i

Huihui Qwen3.8 27B Abliterated AWQ MTP is an open-source language model by shawnw3i. Features: 27b LLM, VRAM: 19.1GB, License: apache-2.0, Quantized, LLM Explorer Score: 0.34.

4-bit Abliterated Awq Base model:quantized:qwen/qwen... Base model:qwen/qwen3.8-27b Conversational Endpoints compatible Image-text-to-text Mtp Quantized Qwen3 5 Region:us Safetensors Sharded Tensorflow Uncensored Vllm

Model Card on HF 🤗: https://huggingface.co/shawnw3i/Huihui-Qwen3.8-27B-abliterated-AWQ-MTP

Huihui Qwen3.8 27B Abliterated AWQ MTP Parameters and Internals

LLM Name	Huihui Qwen3.8 27B Abliterated AWQ MTP
Repository 🤗	https://huggingface.co/shawnw3i/Huihui-Qwen3.8-27B-abliterated-AWQ-MTP
Base Model(s)	huihui-ai/Huihui-Qwen3.8-27B-abliterated Qwen/Qwen3.8-27B huihui-ai/Huihui-Qwen3.8-27B-abliterated Qwen/Qwen3.8-27B
Model Size	27b
Required VRAM	19.1 GB
Updated	2026-08-31
Maintainer	shawnw3i
Model Type	qwen3_5
Model Files	3.2 GB: 1-of-7 3.2 GB: 2-of-7 3.2 GB: 3-of-7 3.2 GB: 4-of-7 1.3 GB: 5-of-7 2.5 GB: 6-of-7 2.5 GB: 7-of-7 0.8 GB
AWQ Quantization	Yes
Quantization Type	awq
Model Architecture	Qwen3_5ForConditionalGeneration
License	apache-2.0
Model Max Length	262144
Transformers Version	5.16.1
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Torch Data Type	float16
Errors	replace

Rank the Huihui Qwen3.8 27B Abliterated AWQ MTP capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to Huihui Qwen3.8 27B Abliterated AWQ MTP

Best Alternatives	Context / RAM	Downloads	Likes
Qwen3.6 27B AWQ INT4	0K / 20.4 GB	2563390	103
Qwen3.8 27B AWQ INT4	0K / 21 GB	21015	32
Qwen3.6 27B AWQ	0K / 22 GB	668924	24
Qwen3.6 27B AWQ BF16 INT4	0K / 28.3 GB	419171	41
Qwen3.8 27B AWQ 4bit	0K / 27.7 GB	11874	6
Qwen3.8 27B AWQ 5.0bpw	0K / 17.3 GB	1445	20
...wen3.6 27B Abliterated AWQ MTP	0K / 18.4 GB	244978	12
Qwen3.8 27B AWQ BF16 INT4	0K / 28.8 GB	1571	11
Qwen3.8 27B W4A16 AWQ	0K / 0.8 GB	3413	7
Qwen3.5 27B AWQ 4bit	0K / 20 GB	279064	42

Note: green Score (e.g. "73.2") means that the model is better than shawnw3i/Huihui-Qwen3.8-27B-abliterated-AWQ-MTP.