Instructions to use UnfilteredAI/Mia-001 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use UnfilteredAI/Mia-001 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="UnfilteredAI/Mia-001")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("UnfilteredAI/Mia-001")
model = AutoModelForCausalLM.from_pretrained("UnfilteredAI/Mia-001")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use UnfilteredAI/Mia-001 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "UnfilteredAI/Mia-001"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "UnfilteredAI/Mia-001",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/UnfilteredAI/Mia-001

SGLang

How to use UnfilteredAI/Mia-001 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "UnfilteredAI/Mia-001" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "UnfilteredAI/Mia-001",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "UnfilteredAI/Mia-001" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "UnfilteredAI/Mia-001",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use UnfilteredAI/Mia-001 with Docker Model Runner:
```
docker model run hf.co/UnfilteredAI/Mia-001
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Model Name: Mia-001

Model Type: Text Generation

Description: Mia-001 is an uncensored AI text generation model developed by MysteriousAI, aimed at advancing and democratizing artificial intelligence through open source and open science initiatives. The model is designed to push the boundaries of creativity and innovation in natural language generation tasks.

Key Features:

Uncensored Text Generation: Mia-001 generates text without censorship, allowing users to explore a wide range of applications without limitations.
Model Size: Mia-001 has 110 million parameters, providing a balance between model complexity and efficiency.
Tensor Type: The model uses FP16 tensor type for efficient computation.
Inference Endpoints: Mia-001 can be loaded on Inference API for serverless deployment, enabling easy integration into applications.

Use Cases:

Content Generation: Mia-001 can be used for generating diverse content, including articles, stories, dialogues, and more.
Creative Writing: Writers and artists can leverage Mia-001 to explore new ideas and narrative structures in their creative works.
Chatbots and Conversational Agents: The model can power chatbots and conversational agents with natural and engaging dialogue generation capabilities.
AI-driven Applications: Mia-001 enables the development of AI-driven applications in areas such as virtual assistants.

Ethical Considerations:

Content Moderation: Users are advised to exercise caution and responsibility when Using Mia-001 in applications involving sensitive or potentially harmful content.
Bias and Fairness: MysteriousAI is committed to addressing biases and promoting fairness in AI models, and ongoing efforts are made to mitigate any biases present in Mia-001.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	29.05
AI2 Reasoning Challenge (25-Shot)	22.78
HellaSwag (10-Shot)	28.02
MMLU (5-Shot)	23.66
TruthfulQA (0-shot)	48.25
Winogrande (5-shot)	51.62
GSM8k (5-shot)	0.00

Downloads last month: 868

Safetensors

Model size

0.1B params

Tensor type

F16

Model tree for UnfilteredAI/Mia-001

Quantizations

4 models

Dataset used to train UnfilteredAI/Mia-001

Spaces using UnfilteredAI/Mia-001 2

Evaluation results

normalized accuracy on AI2 Reasoning Challenge (25-Shot)
test set Open LLM Leaderboard

22.780
normalized accuracy on HellaSwag (10-Shot)
validation set Open LLM Leaderboard

28.020
accuracy on MMLU (5-Shot)
test set Open LLM Leaderboard

23.660
mc2 on TruthfulQA (0-shot)
validation set Open LLM Leaderboard

48.250
accuracy on Winogrande (5-shot)
validation set Open LLM Leaderboard

51.620
accuracy on GSM8k (5-shot)
test set Open LLM Leaderboard

0.000