What are the hardware requirements for Hy3 GPTQ RTN 4bit Tp8?

Hy3 GPTQ RTN 4bit Tp8 requires approximately 161.3 GB of VRAM and supports a context window of 256K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Hy3 GPTQ RTN 4bit Tp8 and how large is it?

Hy3 GPTQ RTN 4bit Tp8 is developed by avtc, a model with 295b 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 Hy3 GPTQ RTN 4bit Tp8?

Hy3 GPTQ RTN 4bit Tp8 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Hy3 GPTQ RTN 4bit Tp8 by avtc — VRAM 161.3GB, 256K context

Name: Hy3 GPTQ RTN 4bit Tp8
Author: avtc

Hy3 GPTQ RTN 4bit Tp8 is an open-source language model by avtc. Features: 295b LLM, VRAM: 161.3GB, Context: 256K, License: apache-2.0, Quantized, LLM Explorer Score: 0.25.

4-bit 4bit Base model:quantized:tencent/h... Base model:tencent/hy3 Conversational Gptq Hunyuan Hy3 Hy v3 Moe Quantization Quantized Region:us Rtn Safetensors Sharded Tensorflow Vllm

Model Card on HF 🤗: https://huggingface.co/avtc/Hy3-GPTQ-RTN-4bit-tp8

Hy3 GPTQ RTN 4bit Tp8 Parameters and Internals

LLM Name	Hy3 GPTQ RTN 4bit Tp8
Repository 🤗	https://huggingface.co/avtc/Hy3-GPTQ-RTN-4bit-tp8
Base Model(s)	Hy3 tencent/Hy3
Model Size	295b
Required VRAM	161.3 GB
Updated	2026-07-09
Maintainer	avtc
Model Type	hy_v3
Model Files	4.3 GB: 1-of-38 4.3 GB: 2-of-38 4.3 GB: 3-of-38 4.3 GB: 4-of-38 4.3 GB: 5-of-38 4.3 GB: 6-of-38 4.3 GB: 7-of-38 4.3 GB: 8-of-38 4.3 GB: 9-of-38 4.3 GB: 10-of-38 4.3 GB: 11-of-38 4.3 GB: 12-of-38 4.3 GB: 13-of-38 4.3 GB: 14-of-38 4.3 GB: 15-of-38 4.3 GB: 16-of-38 4.3 GB: 17-of-38 4.3 GB: 18-of-38 4.3 GB: 19-of-38 4.3 GB: 20-of-38 4.3 GB: 21-of-38 4.3 GB: 22-of-38 4.3 GB: 23-of-38 4.3 GB: 24-of-38 4.3 GB: 25-of-38 4.3 GB: 26-of-38 4.3 GB: 27-of-38 4.3 GB: 28-of-38 4.3 GB: 29-of-38 4.3 GB: 30-of-38 4.3 GB: 31-of-38 4.3 GB: 32-of-38 4.3 GB: 33-of-38 4.3 GB: 34-of-38 4.3 GB: 35-of-38 4.3 GB: 36-of-38 4.3 GB: 37-of-38 2.2 GB: 38-of-38
GPTQ Quantization	Yes
Quantization Type	gptq\|4bit
Model Architecture	HYV3ForCausalLM
License	apache-2.0
Context Length	262144
Model Max Length	262144
Transformers Version	5.9.0
Tokenizer Class	PreTrainedTokenizerFast
Padding Token	<｜hy_pad:opensource｜>
Vocabulary Size	120832

Rank the Hy3 GPTQ RTN 4bit Tp8 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 Hy3 GPTQ RTN 4bit Tp8

Best Alternatives	Context / RAM	Downloads	Likes
Hy3 Preview 4bit	256K / 166.5 GB	1563	3
Hy3 Preview 8bit	256K / 235 GB	141	1
Hy3 Preview 6bit	256K / 227.6 GB	93	1
Hy3 Heretic	256K / 323.1 GB	324	6

Note: green Score (e.g. "73.2") means that the model is better than avtc/Hy3-GPTQ-RTN-4bit-tp8.