You can deploy this model and start earning money today!

LezleeFirestone/
comrade-ai-2

Name: comrade-ai-2
Availability: InStock

safetensors

gguf

mistral

Views

No views yet

comrade-ai-2 LoRA (Mistral-7B-Instruct)

Base model: mistralai/Mistral-7B-Instruct-v0.3
Adapter type: LoRA (PEFT)
Author: Lezlee Firestone Released: 12/9/2025

Updates 12/9/2025

200 more entries added to training data Training data included in new version for review

Purpose

comrade-ai-2 is a LoRA adapter for Mistral 7B Instruct, fine-tuned to answer questions from a Marxist–Leninist anti-revisionist perspective.

It focuses on:

Structural, materialist analysis of capitalism and the state
Clear distinctions between Marxism–Leninism and liberal / social-democratic lines
Concrete, non-academic language where possible
Discipline against tailing bourgeois narratives (“anti-revisionist” focus)

Training data

Private dataset.jsonl with ~500 manually written instruction–response pairs, covering:

Foundations of Marxism–Leninism (dialectical & historical materialism, class struggle, the state, imperialism)
Critiques of common liberal/anarchist/autonomist arguments
Line discipline: avoiding moralizing and idealism; focusing on class and structure
Applied analysis to present-day capitalism and imperialism

Format is Alpaca-style:

json

{"instruction": "...", "input": "", "output": "..."}```


##  How To Use

```from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

base_model = "mistralai/Mistral-7B-Instruct-v0.3"
adapter_id = "LezleeFirestone/comrade-ai-2"

tokenizer = AutoTokenizer.from_pretrained(base_model)
model = AutoModelForCausalLM.from_pretrained(
    base_model,
    device_map="auto",
    torch_dtype="auto"
)

model = PeftModel.from_pretrained(model, adapter_id)

prompt = "Explain the difference between social democracy and Marxism–Leninism."
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=512)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))```



## How To Use With Ollama

Create a ModelFile-
Open a file in notepad++ and save as ModelFile (no extension)

FROM mistral:7b-instruct

ADAPTER https://huggingface.co/LezleeFirestone/comrade-ai-2/resolve/main/adapter_model.safetensors

PARAMETER temperature 0.7 PARAMETER num_ctx 8192```

Then

ollama run comrade-ai-2

License

Base model: see mistralai/Mistral-7B-Instruct-v0.3 license

LoRA weights: Creative Commons Attribution–NonCommercial 4.0 International (CC BY-NC 4.0)

This LoRA is for political education, study, and experimentation. It is not a replacement for real-world organising or reading the primary texts.