What are the hardware requirements for Notux 8x7b V1 GPTQ?

Notux 8x7b V1 GPTQ requires approximately 23.8 GB of VRAM and supports a context window of 32K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Notux 8x7b V1 GPTQ and how large is it?

Notux 8x7b V1 GPTQ is developed by TheBloke, a model with 46.7b 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 Notux 8x7b V1 GPTQ?

Notux 8x7b V1 GPTQ is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Notux 8x7b V1 GPTQ by TheBloke — VRAM 23.8GB, 32K context

Name: Notux 8x7b V1 GPTQ
Author: TheBloke

Notux 8x7b V1 GPTQ is an open-source language model by TheBloke. Features: 46.7b LLM, VRAM: 23.8GB, Context: 32K, License: apache-2.0, MoE, Quantized, LLM Explorer Score: 0.1.

4-bit Base model:argilla/notux-8x7b-... Base model:quantized:argilla/n... Conversational Dataset:argilla/ultrafeedback-... De Dpo En Es Fr Gptq It Mixtral Moe Preference Quantized Region:us Rlaif Safetensors Ultrafeedback

Model Card on HF 🤗: https://huggingface.co/TheBloke/notux-8x7b-v1-GPTQ

Notux 8x7b V1 GPTQ Parameters and Internals

Model Type

Pretrained generative Sparse Mixture of Experts

Additional Notes

This model represents the first experiment in tuning a MoE model that has already been fine-tuned with DPO.

Supported Languages

English (NLP), Spanish (NLP), Italian (NLP), German (NLP), French (NLP)

Training Details

Data Sources:

argilla/ultrafeedback-binarized-preferences-cleaned

Methodology:

DPO (Direct Preference Optimization)

Training Time:

~10hr

Hardware Used:

8 x H100 80GB hosted in runpod.io

LLM Name	Notux 8x7b V1 GPTQ
Repository 🤗	https://huggingface.co/TheBloke/notux-8x7b-v1-GPTQ
Model Name	Notux 8X7B v1
Model Creator	Argilla
Base Model(s)	Notux 8x7b V1 argilla/notux-8x7b-v1
Model Size	46.7b
Required VRAM	23.8 GB
Updated	2026-08-01
Maintainer	TheBloke
Model Type	mixtral
Model Files	23.8 GB
Supported Languages	en de es fr it
GPTQ Quantization	Yes
Quantization Type	gptq
Model Architecture	MixtralForCausalLM
License	apache-2.0
Context Length	32768
Model Max Length	32768
Transformers Version	4.37.0.dev0
Tokenizer Class	LlamaTokenizer
Padding Token	</s>
Vocabulary Size	32000
Torch Data Type	bfloat16

Rank the Notux 8x7b V1 GPTQ 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 Notux 8x7b V1 GPTQ

Best Alternatives	Context / RAM	Downloads	Likes
Mixtral 8x7B V0.1 GPTQ	32K / 23.8 GB	181	126
...ixtral 8x7B Instruct V0.1 GPTQ	32K / 23.8 GB	210	141
...Hermes 2 Mixtral 8x7B DPO GPTQ	32K / 23.8 GB	31	26
Dolphin 2.5 Mixtral 8x7b GPTQ	32K / 23.8 GB	81	114
...Hermes 2 Mixtral 8x7B SFT GPTQ	32K / 23.8 GB	10	11
Bagel DPO 8x7b V0.2 GPTQ	32K / 23.8 GB	12	2
Open Gpt4 8x7B V0.2 GPTQ	32K / 23.8 GB	9	6
...xtral Instruct 8x7b Zloss GPTQ	32K / 23.8 GB	7	2
Sensualize Mixtral GPTQ	32K / 23.8 GB	9	5
....1 LimaRP ZLoss DARE TIES GPTQ	32K / 23.8 GB	5	6

Note: green Score (e.g. "73.2") means that the model is better than TheBloke/notux-8x7b-v1-GPTQ.