You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Mistral-7B-Discord

What is this

A model finetuned on over 236k discord messages over the course of 869 days.

Goals

This model was trained to capture the tone, quirks, and conversational style of a long-term discord exchange between close friends. It aims to replicate casual, often chaotic, sometimes deep, and context-aware banter including inside jokes, emotional shifts, and rapid topic switching.

Use cases

Casual chatbots with personality
Character-driven AI

Limitations

Highly specialized: This is a small (7B) model fine-tuned on niche data. It won't generalize well and sometimes says dumb stuff.
Informal: Expect slang, memes, inconsistent grammar, and emotionally charged responses.

How to use

huggingface-cli login

Then:

ollama run hf.co/NotClavilux/mistral-7b-discord:Q4_K_M

Or just use llama.cpp or whatever you want.

Credits

Trained on data collected with consent. Project by NotClavilux.

Downloads last month: -

GGUF

Model size

7B params

Architecture

llama

Hardware compatibility

4-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for NotClavilux/mistral-7b-discord

Base model

mistralai/Mistral-7B-v0.3

Finetuned

mistralai/Mistral-7B-Instruct-v0.3

Quantized

(212)

this model