Instructions to use ayoubkirouane/Mistral-Depth-UP-Scaled-9B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ayoubkirouane/Mistral-Depth-UP-Scaled-9B with Transformers:

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

pipe = pipeline("text-generation", model="ayoubkirouane/Mistral-Depth-UP-Scaled-9B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ayoubkirouane/Mistral-Depth-UP-Scaled-9B")
model = AutoModelForCausalLM.from_pretrained("ayoubkirouane/Mistral-Depth-UP-Scaled-9B")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ayoubkirouane/Mistral-Depth-UP-Scaled-9B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ayoubkirouane/Mistral-Depth-UP-Scaled-9B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ayoubkirouane/Mistral-Depth-UP-Scaled-9B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ayoubkirouane/Mistral-Depth-UP-Scaled-9B

SGLang

How to use ayoubkirouane/Mistral-Depth-UP-Scaled-9B 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 "ayoubkirouane/Mistral-Depth-UP-Scaled-9B" \
    --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": "ayoubkirouane/Mistral-Depth-UP-Scaled-9B",
		"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 "ayoubkirouane/Mistral-Depth-UP-Scaled-9B" \
        --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": "ayoubkirouane/Mistral-Depth-UP-Scaled-9B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ayoubkirouane/Mistral-Depth-UP-Scaled-9B with Docker Model Runner:
```
docker model run hf.co/ayoubkirouane/Mistral-Depth-UP-Scaled-9B
```

Mistral-Depth-UP-Scaled-9B

An auto-regressive causal LM created by combining 2x finetuned mistral 7B into one.

Benchmarks

Coming soon.

Usage :

# Load in4Bit

from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
import torch

nf4_config = BitsAndBytesConfig(
   load_in_4bit=True,
   bnb_4bit_quant_type="nf4",
   bnb_4bit_use_double_quant=True,
   bnb_4bit_compute_dtype=torch.bfloat16
)
model = AutoModelForCausalLM.from_pretrained(
    "ayoubkirouane/Mistral-Depth-UP-Scaled-9B",
    device_map='auto',
    quantization_config=nf4_config,
    use_cache=False
)
tokenizer = AutoTokenizer.from_pretrained("ayoubkirouane/Mistral-Depth-UP-Scaled-9B")

tokenizer.pad_token = tokenizer.eos_token
tokenizer.padding_side = "right"

def generate_response(prompt, model , max_new_tokens):
  encoded_input = tokenizer(prompt,  return_tensors="pt", add_special_tokens=True)
  model_inputs = encoded_input.to('cuda')

  generated_ids = model.generate(**model_inputs, max_new_tokens=max_new_tokens, do_sample=True, pad_token_id=tokenizer.eos_token_id)

  decoded_output = tokenizer.batch_decode(generated_ids)

  return decoded_output[0].replace(prompt, "")


generate_response(prompt="What is GANs ?", model=model , max_new_tokens=100)

Downloads last month: 5

Safetensors

Model size

9B params

Tensor type

BF16