What are the hardware requirements for Deepsex 34B 4bpw H8 EXL2?

Deepsex 34B 4bpw H8 EXL2 requires approximately 18.1 GB of VRAM and supports a context window of 4K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Deepsex 34B 4bpw H8 EXL2 and how large is it?

Deepsex 34B 4bpw H8 EXL2 is developed by Redwood0, a model with 34b 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 Deepsex 34B 4bpw H8 EXL2?

Deepsex 34B 4bpw H8 EXL2 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Deepsex 34B 4bpw H8 EXL2 by Redwood0 — VRAM 18.1GB, 4K context

Name: Deepsex 34B 4bpw H8 EXL2
Author: Redwood0

Deepsex 34B 4bpw H8 EXL2 is an open-source language model by Redwood0. Features: 34b LLM, VRAM: 18.1GB, Context: 4K, License: mit, Quantized, LLM Explorer Score: 0.1.

Custom code Dataset:lemonilia/limarp Dataset:pygmalionai/pippa En Exl2 Feature-extraction Not-for-all-audiences Pytorch Quantized Region:us Roleplay Sharded Tensorflow Yi

Model Card on HF 🤗: https://huggingface.co/Redwood0/deepsex-34b-4bpw-h8-exl2

Deepsex 34B 4bpw H8 EXL2 Parameters and Internals

Model Type

text generation, role-play

Additional Notes

Model is proficient in role-playing skills and handles NSFW content well.

Supported Languages

en (proficient)

Training Details

Data Sources:

light novels, nsfw novels, limarp+pippa dataset

Data Volume:

4GB of various light novels

Methodology:

performed similarity deduplication, used qlora for fine-tuning, cleaned and scored Q&A pairs

Context Length:

4096

Input Output

Input Format:

alpaca format

Output Format:

role-playing responses

LLM Name	Deepsex 34B 4bpw H8 EXL2
Repository 🤗	https://huggingface.co/Redwood0/deepsex-34b-4bpw-h8-exl2
Model Size	34b
Required VRAM	18.1 GB
Updated	2026-08-08
Maintainer	Redwood0
Model Type	Yi
Model Files	8.5 GB: 1-of-3 8.5 GB: 2-of-3 1.1 GB: 3-of-3
Supported Languages	en
Quantization Type	exl2
Model Architecture	YiForCausalLM
License	mit
Context Length	4096
Model Max Length	4096
Transformers Version	4.34.1
Tokenizer Class	YiTokenizer
Padding Token	<unk>
Vocabulary Size	64000
Torch Data Type	bfloat16

Rank the Deepsex 34B 4bpw H8 EXL2 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 Deepsex 34B 4bpw H8 EXL2

Best Alternatives	Context / RAM	Downloads	Likes
...lpaca Rpv3 Scipy 4bpw Hb6 EXL2	195K / 18.7 GB	10	2
...200K Alpaca Rpv3 4bpw Hb6 EXL2	195K / 18.7 GB	2	2
Yi 34B 200K 4.0bpw H6 EXL2	195K / 18.1 GB	4	3
Yi 34B 200K 4.65bpw H6 EXL2	195K / 20.8 GB	5	6
Yi 34B 200K 8.0bpw H8 EXL2	195K / 34.8 GB	2	3
Yi 34B 200K 5.0bpw H6 EXL2	195K / 22.3 GB	4	1
Deepsex 34B 4bpw H6 EXL2	4K / 18.1 GB	2	4
Yi 34B 5.0bpw H6 EXL2	4K / 22.2 GB	11	3
Yi 34B 4.0bpw H6 EXL2	4K / 18.1 GB	6	1
Yi 34B 8.0bpw H8 EXL2	4K / 34.9 GB	2	1