What are the hardware requirements for Solstice 11B V1.8.0bpw H8 EXL2?

Solstice 11B V1.8.0bpw H8 EXL2 requires approximately 10.9 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 Solstice 11B V1.8.0bpw H8 EXL2 and how large is it?

Solstice 11B V1.8.0bpw H8 EXL2 is developed by LoneStriker, a model with 11b 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 Solstice 11B V1.8.0bpw H8 EXL2?

Solstice 11B V1.8.0bpw H8 EXL2 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Solstice 11B V1.8.0bpw H8 EXL2 by LoneStriker — VRAM 10.9GB, 4K context

Name: Solstice 11B V1.8.0bpw H8 EXL2
Author: LoneStriker

Solstice 11B V1.8.0bpw H8 EXL2 is an open-source language model by LoneStriker. Features: 11b LLM, VRAM: 10.9GB, Context: 4K, License: cc-by-nc-4.0, Quantized, LLM Explorer Score: 0.11.

Dataset:himitsui/lewd-assistan... En Endpoints compatible Exl2 Llama Pytorch Quantized Region:us Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/LoneStriker/Solstice-11B-v1-8.0bpw-h8-exl2

Solstice 11B V1.8.0bpw H8 EXL2 Parameters and Internals

Model Type

NSFW

Use Cases

Areas:

NSFW

Limitations:

May speak as {{user}}, May be inconsistent, Steers towards NSFW

Additional Notes

Experimental model with issues such as inconsistent outputs and rambling

Supported Languages

en (primary)

Training Details

Data Sources:

Himitsui/Lewd-Assistant-v1

Methodology:

Finetuned off several instruct datasets that are NSFW

Input Output

Input Format:

Alpaca

Accepted Modalities:

text

LLM Name	Solstice 11B V1.8.0bpw H8 EXL2
Repository 🤗	https://huggingface.co/LoneStriker/Solstice-11B-v1-8.0bpw-h8-exl2
Model Size	11b
Required VRAM	10.9 GB
Updated	2026-05-24
Maintainer	LoneStriker
Model Type	llama
Model Files	8.6 GB: 1-of-2 2.3 GB: 2-of-2
Supported Languages	en
Quantization Type	exl2
Model Architecture	LlamaForCausalLM
License	cc-by-nc-4.0
Context Length	4096
Model Max Length	4096
Transformers Version	4.38.0.dev0
Tokenizer Class	LlamaTokenizer
Padding Token	</s>
Vocabulary Size	32000
Torch Data Type	bfloat16

Rank the Solstice 11B V1.8.0bpw 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 Solstice 11B V1.8.0bpw H8 EXL2

Best Alternatives	Context / RAM	Downloads	Likes
... 11B V2.5 Instruct FP8 Dynamic	32K / 11.5 GB	33	2
...istral 11B Omni OPA U1k Ver0.7	32K / 21.4 GB	12	0
Mistral 11B OP U1k Ver0.7	32K / 21.4 GB	12	0
Mistral 11B Omni OP U1k Ver0.5	32K / 21.4 GB	19	0
...istral 11B V2 HQQ 4bit Smashed	8K / 6 GB	5	0
...mbulvetr 11B V2.8.0bpw H8 EXL2	4K / 10.9 GB	6	7
...mbulvetr 11B V2.6.0bpw H6 EXL2	4K / 8.2 GB	3	4
...mbulvetr 11B V2.3.0bpw H6 EXL2	4K / 4.3 GB	6	1
...mbulvetr 11B V2.4.0bpw H6 EXL2	4K / 5.6 GB	5	0
A6.1	1024K / 16.1 GB	130	0

Note: green Score (e.g. "73.2") means that the model is better than LoneStriker/Solstice-11B-v1-8.0bpw-h8-exl2.