Instructions to use mymaia/Magiq-Core-4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mymaia/Magiq-Core-4 with Transformers:

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

pipe = pipeline("text-generation", model="mymaia/Magiq-Core-4")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mymaia/Magiq-Core-4")
model = AutoModelForCausalLM.from_pretrained("mymaia/Magiq-Core-4")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use mymaia/Magiq-Core-4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mymaia/Magiq-Core-4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mymaia/Magiq-Core-4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/mymaia/Magiq-Core-4

SGLang

How to use mymaia/Magiq-Core-4 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 "mymaia/Magiq-Core-4" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mymaia/Magiq-Core-4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "mymaia/Magiq-Core-4" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mymaia/Magiq-Core-4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use mymaia/Magiq-Core-4 with Docker Model Runner:
```
docker model run hf.co/mymaia/Magiq-Core-4
```

Model Card for Magiq Core 4

Features of the Magiq Core Model Development and Purpose:
The Magiq Core 4 model represents a significant advancement in proprietary model development for MAIA. Trained with an extensive dataset of 400,000 examples, compared to the 50,000 examples used for Magiq Core 0, this new version substantially improves performance. The primary purpose of Magiq Core 4 is to enhance interactions in English, French, and Italian, each with unique linguistic peculiarities and nuances.

Importance of Language Specificity:
Magiq Core 4 focuses on the specificities of English, French, and Italian languages and their respective cultural nuances. This attention is crucial in a field where most AI models are predominantly based on English datasets, which differ in structure and vocabulary from Romance languages like French and Italian.

Benefits of Language Specialization:
Greater Accuracy: Magiq Core 4 offers a more accurate understanding and generation of responses, respecting each language's nuances. Improved Contextual Understanding: Facilitates more fluid and natural interactions by interpreting linguistic and cultural context. Enhanced Personalization: This approach allows for responses tailored to user preferences and needs. Specialization by Device and Dialogic Tools: Magiq Core 4 is also specialized for varying outputs depending on the device and is optimized for interactions with dialogic tools like copilots and virtual assistants, ensuring effective communication across different devices.

In conclusion, Magiq Core 4 was developed to create a more understandable, customizable, and accessible AI system, paving the way for a future where AI is more integrated into daily life.

Downloads last month: 12

Safetensors

Model size

7B params

Tensor type

BF16

Model tree for mymaia/Magiq-Core-4

Base model

mistralai/Mistral-7B-Instruct-v0.2

Finetuned

(1088)

this model

Collection including mymaia/Magiq-Core-4

Magiq

Collection

1 item • Updated Apr 8, 2024 • 2