Spaces:
Runtime error
Runtime error
A newer version of the Gradio SDK is available:
5.5.0
metadata
title: Prometh-MOEM-V.01 Model Showcase
emoji: π
colorFrom: red
colorTo: pink
sdk: gradio
pinned: false
license: apache-2.0
language:
- en
Prometh-MOEM-V.01 Model Card π
Prometh-MOEM-V.01 is a pioneering Mixture of Experts (MoE) model, blending the capabilities of multiple foundational models to enhance performance across a variety of tasks. This model leverages the collective strengths of its components, achieving unparalleled accuracy, speed, and versatility.
π Model Sources and Components
This MoE model amalgamates specialized models including:
- Wtzwho/Prometh-merge-test2
- mistralai/Mistral-7B-Instruct-v0.2
- Wtzwho/Prometh-merge-test3
- meta-math/MetaMath-Mistral-7B
π Key Features
- Enhanced Performance: Tailored for peak accuracy and efficiency.
- Versatility: Exceptionally adaptable across a wide range of NLP tasks.
- State-of-the-Art Integration: Incorporates the latest in AI research for effective model integration.
π Application Areas
Prometh-MOEM-V.01 excels in:
- Text generation
- Sentiment analysis
- Language translation
- Question answering
π» Usage Instructions
To utilize Prometh-MOEM-V.01 in your projects:
pip install -qU transformers bitsandbytes accelerate
from transformers import AutoTokenizer, pipeline
import torch
model = "Wtzwho/Prometh-MOEM-V.01"
tokenizer = AutoTokenizer.from_pretrained(model)
# Setup pipeline
pipeline = pipeline(
"text-generation",
model=model,
model_kwargs={"torch_dtype": torch.float16, "load_in_4bit": True},
)
# Example query
messages = [{"role": "user", "content": "Explain what a Mixture of Experts is in less than 100 words."}]
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])