Instructions to use mymaia/Magiq-3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mymaia/Magiq-3 with Transformers:

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

pipe = pipeline("text-generation", model="mymaia/Magiq-3")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mymaia/Magiq-3")
model = AutoModelForCausalLM.from_pretrained("mymaia/Magiq-3")
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]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use mymaia/Magiq-3 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mymaia/Magiq-3"
# 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-3",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/mymaia/Magiq-3

SGLang

How to use mymaia/Magiq-3 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-3" \
    --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-3",
		"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-3" \
        --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-3",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use mymaia/Magiq-3 with Docker Model Runner:
```
docker model run hf.co/mymaia/Magiq-3
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Model Card for Magiq 3

Magiq 3 as a Mixture of Experts (MoE)

The MoE architecture of Magiq 3 combines the specialized capabilities of MAGIQ Core-0, MAGIQ Translator-0, and MAGIQ Logic-0 into a cohesive, intelligent framework.

This structure enables MAIA to offer unparalleled assistance, characterized by deep understanding, linguistic flexibility, and logical reasoning. Magiq3's MoE design not only optimizes performance across different tasks but also ensures that MAIA's interactions are as human-like and natural as possible, catering to a wide range of user needs and preferences.

Downloads last month: 7

Safetensors

Model size

19B params

Tensor type

BF16

Model tree for mymaia/Magiq-3

Base model

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

Finetuned

(1088)

this model

Quantizations

3 models