What are the hardware requirements for Qwen3.5 397B A17B Heretic Int4 AutoRound?

Qwen3.5 397B A17B Heretic Int4 AutoRound requires approximately 212.3 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Qwen3.5 397B A17B Heretic Int4 AutoRound and how large is it?

Qwen3.5 397B A17B Heretic Int4 AutoRound is developed by happypatrick, a model with 397b 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 Qwen3.5 397B A17B Heretic Int4 AutoRound?

Qwen3.5 397B A17B Heretic Int4 AutoRound is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Qwen3.5 397B A17B Heretic Int4 AutoRound by happypatrick — VRAM 212.3GB

Name: Qwen3.5 397B A17B Heretic Int4 AutoRound
Author: happypatrick

Qwen3.5 397B A17B Heretic Int4 AutoRound is an open-source language model by happypatrick. Features: 397b LLM, VRAM: 212.3GB, License: apache-2.0, LLM Explorer Score: 0.27.

Arxiv:2309.05516 4-bit Abliterated Ar Auto-round Autoround Base model:quantized:radicalno... Base model:radicalnotionai/qwe... Code Conversational De En Es Fr Gptq Heretic Int4 Ja Ko Moe Pt Quantization Qwen3 5 moe Reasoning Region:us Ru Safetensors Sharded Tensorflow Th Uncensored Vi Vision Zh

Model Card on HF 🤗: https://huggingface.co/happypatrick/Qwen3.5-397B-A17B-heretic-int4-AutoRound

Qwen3.5 397B A17B Heretic Int4 AutoRound Parameters and Internals

LLM Name	Qwen3.5 397B A17B Heretic Int4 AutoRound
Repository 🤗	https://huggingface.co/happypatrick/Qwen3.5-397B-A17B-heretic-int4-AutoRound
Base Model(s)	trohrbaugh/Qwen3.5-397B-A17B-heretic trohrbaugh/Qwen3.5-397B-A17B-heretic
Model Size	397b
Required VRAM	212.3 GB
Updated	2026-08-07
Maintainer	happypatrick
Model Type	qwen3_5_moe
Model Files	5.4 GB: 1-of-40 5.4 GB: 2-of-40 5.4 GB: 3-of-40 5.4 GB: 4-of-40 5.4 GB: 5-of-40 5.4 GB: 6-of-40 5.4 GB: 7-of-40 5.4 GB: 8-of-40 5.4 GB: 9-of-40 5.4 GB: 10-of-40 5.4 GB: 11-of-40 5.4 GB: 12-of-40 5.4 GB: 13-of-40 5.4 GB: 14-of-40 5.4 GB: 15-of-40 5.4 GB: 16-of-40 5.4 GB: 17-of-40 5.4 GB: 18-of-40 5.4 GB: 19-of-40 5.4 GB: 20-of-40 5.4 GB: 21-of-40 5.4 GB: 22-of-40 5.4 GB: 23-of-40 5.4 GB: 24-of-40 5.4 GB: 25-of-40 5.4 GB: 26-of-40 5.4 GB: 27-of-40 5.4 GB: 28-of-40 5.4 GB: 29-of-40 5.4 GB: 30-of-40 5.4 GB: 31-of-40 5.4 GB: 32-of-40 5.4 GB: 33-of-40 5.4 GB: 34-of-40 5.4 GB: 35-of-40 5.4 GB: 36-of-40 5.4 GB: 37-of-40 5.4 GB: 38-of-40 5.1 GB: 39-of-40 2.0 GB: 40-of-40
Supported Languages	en zh ja ko fr de es pt ru ar th vi
Model Architecture	Qwen3_5MoeForConditionalGeneration
License	apache-2.0
Model Max Length	262144
Transformers Version	5.2.0
Tokenizer Class	TokenizersBackend
Padding Token	<\|endoftext\|>
Errors	replace

Rank the Qwen3.5 397B A17B Heretic Int4 AutoRound 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 Qwen3.5 397B A17B Heretic Int4 AutoRound

Best Alternatives	Context / RAM	Downloads	Likes
Qwen3.5 397B A17B	0K / 369.8 GB	331990	1549
Qwen3.5 397B A17B NVFP4	0K / 251.1 GB	428565	102
Qwen3.5 397B A17B NVFP4 V2	0K / 243.5 GB	23781	5
Ornith 1.0 397B	0K / 288.5 GB	318396	260
Ornith 1.0 397B FP8	0K / 147.7 GB	597119	183
XYZ Aquila Pro	0K / 225 GB	1805	375
Qwen3.5 397B A17B	0K / 369.8 GB	53	13
Qwen3.5 397B A17B FP8	0K / 184.9 GB	825515	182
Qwen3.5 397B A17B MXFP4	0K / 98.9 GB	21989	6
Ornith 1.0 397B	0K / 288.5 GB	447	4

Note: green Score (e.g. "73.2") means that the model is better than happypatrick/Qwen3.5-397B-A17B-heretic-int4-AutoRound.