Instructions to use MoxoffAdmin/Mistral_InfoSynth with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MoxoffAdmin/Mistral_InfoSynth with Transformers:

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("MoxoffAdmin/Mistral_InfoSynth")
model = AutoModelForCausalLM.from_pretrained("MoxoffAdmin/Mistral_InfoSynth")
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 MoxoffAdmin/Mistral_InfoSynth with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/MoxoffAdmin/Mistral_InfoSynth

SGLang

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

Docker Model Runner
How to use MoxoffAdmin/Mistral_InfoSynth with Docker Model Runner:
```
docker model run hf.co/MoxoffAdmin/Mistral_InfoSynth
```

What is the base model used for this one?

by luigisaetta - opened Jan 13, 2024

Discussion

luigisaetta

Jan 13, 2024

Interesting effort. My question is about what is the foundation model used (MIstral 7B?) because in your LinkedIn post I see mentioned a context widow of 8K, but in the model card you talk about Mistra_ita that declares a context widow of 4K.

luigisaetta

Jan 13, 2024

•

edited Jan 13, 2024

ok sorry, I see. It is based on your Mistral_Ita, based on Mistral 7B who has a context window of 8K. I see, my mistake.

MoxoffAdmin

Owner Jan 15, 2024

yes exactly!

MoxoffAdmin changed discussion status to closed Jan 15, 2024

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment