What are the hardware requirements for Tor 8B?

Tor 8B requires approximately 16.8 GB of VRAM and supports a context window of 16K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Tor 8B and how large is it?

Tor 8B is developed by Delta-Vector, a model with 8b 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 Tor 8B?

Tor 8B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Tor 8B by Delta-Vector — VRAM 16.8GB, 16K context

Name: Tor 8B
Author: Delta-Vector

Tor 8B is an open-source language model by Delta-Vector. Features: 8b LLM, VRAM: 16.8GB, Context: 16K, License: agpl-3.0, Instruction-Based, LLM Explorer Score: 0.2.

Base model:finetune:nvidia/mis... Base model:nvidia/mistral-nemo... Chat Dataset:anthracite-org/c2 logs... Dataset:anthracite-org/kalo-op... Dataset:anthracite-org/kalo mi... Dataset:anthracite-org/kalo op... Dataset:anthracite-org/nopm cl... Dataset:epiculous/synthrp-gens... Dataset:epiculous/synthstruct-... Dataset:lodrick-the-lafted/kal... En Instruct Mistral Model-index Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/Delta-Vector/Tor-8B

Tor 8B Parameters and Internals

Model Type

Text Generation

Additional Notes

Inspired by Darkens-8B and further fine-tuned.

Training Details

Data Sources:

PRIVATE CLAUDE LOG FILTER, anthracite-org/kalo-opus-instruct-22k-no-refusal, Epiculous/SynthRP-Gens-v1.1-Filtered-n-Cleaned, lodrick-the-lafted/kalo-opus-instruct-3k-filtered, anthracite-org/nopm_claude_writing_fixed, Epiculous/Synthstruct-Gens-v1.1-Filtered-n-Cleaned, anthracite-org/kalo_opus_misc_240827, anthracite-org/kalo_misc_part2

Methodology:

Fine-tuning with ChatML formatting

Context Length:

16384

Hardware Used:

10 x Nvidia A40 GPUs

Input Output

Input Format:

ChatML formatting

Performance Tips:

Use Sao10k's Euryale System prompt or 'Roleplay Simple' system prompt provided within SillyTavern.

LLM Name	Tor 8B
Repository 🤗	https://huggingface.co/Delta-Vector/Tor-8B
Base Model(s)	Mistral NeMo Minitron 8B Base nvidia/Mistral-NeMo-Minitron-8B-Base
Model Size	8b
Required VRAM	16.8 GB
Updated	2026-08-25
Maintainer	Delta-Vector
Model Type	mistral
Instruction-Based	Yes
Model Files	4.9 GB: 1-of-4 5.0 GB: 2-of-4 4.9 GB: 3-of-4 2.0 GB: 4-of-4 0.0 GB 0.0 GB
Supported Languages	en
Model Architecture	MistralForCausalLM
License	agpl-3.0
Context Length	16384
Model Max Length	16384
Transformers Version	4.45.0.dev0
Tokenizer Class	PreTrainedTokenizerFast
Padding Token	<pad>
Vocabulary Size	131072
Torch Data Type	bfloat16

Rank the Tor 8B 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 Tor 8B

Best Alternatives	Context / RAM	Downloads	Likes
Ministral 8B Instruct 2410 HF	32K / 32 GB	21	12
Ministrations 8B V1	32K / 16.1 GB	20	29
...l 8B Instruct 2410 Abliterated	32K / 16.1 GB	46	3
Ministral 8B Slerp	32K / 29.2 GB	7	0
Forgotten Safeword 8B V2.2	32K / 16.1 GB	4	1
DeepOpus 1 8B Preview	32K / 16.1 GB	75	2
DeepNeo 1 8B Preview	32K / 16.1 GB	63	2
Bigger Body 8B	32K / 16.1 GB	11	7
Reflect Single Mini8B SftT12	32K / 16.1 GB	5	0
...ct Mini8B MistlrgOrcl460kSftT2	32K / 16.1 GB	5	0

Note: green Score (e.g. "73.2") means that the model is better than Delta-Vector/Tor-8B.