What are the hardware requirements for Qwen3.8 27B Abliterated GPTQ Int4 MTP?

Qwen3.8 27B Abliterated GPTQ Int4 MTP requires approximately 19.6 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Qwen3.8 27B Abliterated GPTQ Int4 MTP and how large is it?

Qwen3.8 27B Abliterated GPTQ Int4 MTP is developed by ericmey, 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 Qwen3.8 27B Abliterated GPTQ Int4 MTP?

Qwen3.8 27B Abliterated GPTQ Int4 MTP is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Qwen3.8 27B Abliterated GPTQ Int4 MTP by ericmey — VRAM 19.6GB

Name: Qwen3.8 27B Abliterated GPTQ Int4 MTP
Author: ericmey

Qwen3.8 27B Abliterated GPTQ Int4 MTP is an open-source language model by ericmey. Features: 27b LLM, VRAM: 19.6GB, License: apache-2.0, Quantized, LLM Explorer Score: 0.3.

4-bit Abliterated Base model:huihui-ai/huihui-qw... Base model:quantized:huihui-ai... Conversational En Endpoints compatible Gptq Image-text-to-text Int4 Intel-arc Mtp Quantized Qwen3 Qwen3 5 Region:us Safetensors Sharded Tensorflow Uncensored Vision Vllm W4a16 Xpu Zh

Model Card on HF 🤗: https://huggingface.co/ericmey/Qwen3.8-27B-abliterated-GPTQ-Int4-MTP

Qwen3.8 27B Abliterated GPTQ Int4 MTP Parameters and Internals

LLM Name	Qwen3.8 27B Abliterated GPTQ Int4 MTP
Repository 🤗	https://huggingface.co/ericmey/Qwen3.8-27B-abliterated-GPTQ-Int4-MTP
Base Model(s)	huihui-ai/Huihui-Qwen3.8-27B-abliterated huihui-ai/Huihui-Qwen3.8-27B-abliterated
Model Size	27b
Required VRAM	19.6 GB
Updated	2026-10-08
Maintainer	ericmey
Model Type	qwen3_5
Model Files	3.5 GB: 1-of-5 4.3 GB: 2-of-5 4.3 GB: 3-of-5 4.3 GB: 4-of-5 3.2 GB: 5-of-5
Supported Languages	en zh
GPTQ Quantization	Yes
Quantization Type	gptq
Model Architecture	Qwen3_5ForConditionalGeneration
License	apache-2.0
Model Max Length	262144
Transformers Version	5.18.0
Tokenizer Class	Qwen2TokenizerFast
Padding Token	<\|endoftext\|>
Errors	replace

Rank the Qwen3.8 27B Abliterated GPTQ Int4 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 Qwen3.8 27B Abliterated GPTQ Int4 MTP

Best Alternatives	Context / RAM	Downloads	Likes
Qwen3.6 27B GPTQ Pro 4bit	0K / 18.8 GB	415407	39
Qwen3.8 27B GPTQ INT4 FP8KV	0K / 20.2 GB	6488	2
Qwen3.5 27B GPTQ Int4	0K / 30.1 GB	105381	57
...en3.8 27B W4A16 AutoRound GPTQ	0K / 19.2 GB	3748	5
Qwen3.8 27B GPTQ 4bit	0K / 21 GB	2433	5
Qwen3.8 27B GPTQ Int4	0K / 20.2 GB	2267	3
Qwen3.8 27B GPTQ INT4 G128	0K / 19.6 GB	2048	3
...7B GPTQ Int4 Sym G128 MTP BF16	0K / 19.6 GB	175	3
... GPTQ Int4 Baked V1 Embed Int8	0K / 16.2 GB	39	4
... V2 GPTQ W4A16 Embed Int4 24GB	0K / 0.8 GB	91	1

Note: green Score (e.g. "73.2") means that the model is better than ericmey/Qwen3.8-27B-abliterated-GPTQ-Int4-MTP.