Instructions to use mistralai/Mistral-7B-Instruct-v0.2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mistralai/Mistral-7B-Instruct-v0.2 with Transformers:

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

pipe = pipeline("text-generation", model="mistralai/Mistral-7B-Instruct-v0.2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-Instruct-v0.2")
model = AutoModelForCausalLM.from_pretrained("mistralai/Mistral-7B-Instruct-v0.2")
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 Settings

vLLM

How to use mistralai/Mistral-7B-Instruct-v0.2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Install mistral-common:
pip install --upgrade mistral-common
# Start the vLLM server:
vllm serve "mistralai/Mistral-7B-Instruct-v0.2" --tokenizer_mode mistral --config_format mistral --load_format mistral --tool-call-parser mistral --enable-auto-tool-choice
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mistralai/Mistral-7B-Instruct-v0.2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/mistralai/Mistral-7B-Instruct-v0.2

SGLang

How to use mistralai/Mistral-7B-Instruct-v0.2 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 "mistralai/Mistral-7B-Instruct-v0.2" \
    --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": "mistralai/Mistral-7B-Instruct-v0.2",
		"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 "mistralai/Mistral-7B-Instruct-v0.2" \
        --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": "mistralai/Mistral-7B-Instruct-v0.2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use mistralai/Mistral-7B-Instruct-v0.2 with Docker Model Runner:
```
docker model run hf.co/mistralai/Mistral-7B-Instruct-v0.2
```

sliding_window appears to be None. TypeError: bad operand type for unary -: 'NoneType'

#56

by narai - opened Feb 29, 2024

Discussion

narai

Feb 29, 2024

Error when running the model card code:

File .../lib/python3.10/site-packages/transformers/models/mistral/modeling_mistral.py:88, in _make_sliding_window_causal_mask(input_ids_shape, dtype, device, past_key_values_length, sliding_window)
86 mask = torch.tril(tensor, diagonal=0)
87 # make the mask banded to account for sliding window
---> 88 mask = torch.triu(mask, diagonal=-sliding_window)
89 mask = torch.log(mask).to(dtype)
91 if past_key_values_length > 0:

TypeError: bad operand type for unary -: 'NoneType'

narai

Feb 29, 2024

Solved with:

ckpt = "mistralai/Mistral-7B-Instruct-v0.2"

config = AutoConfig.from_pretrained(ckpt)
config.update({'sliding_window': 4096})

model = AutoModelForCausalLM.from_pretrained(ckpt, device_map="auto", config=config)
tokenizer = AutoTokenizer.from_pretrained(ckpt, config=config)

ybelkada

Mar 4, 2024

Hi @narai
You can also solve the issue by updating transformers - pip install -U transformers

GTimothee

Mar 22, 2024

you can also just add the option in from_pretrained (instead of using AutoConfig just for that aim): AutoModelForCausalLM.from_pretrained(ckpt, device_map="auto", sliding_window=4096)

deleted

Apr 18, 2024

you can also just add the option in from_pretrained (instead of using AutoConfig just for that aim): AutoModelForCausalLM.from_pretrained(ckpt, device_map="auto", sliding_window=4096)

However it says that v0.2 doesn't use sliding-window-attention, should we set sliding_window=4096, or just set sliding_window=32k

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