Instructions to use tiiuae/falcon-7b-instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tiiuae/falcon-7b-instruct with Transformers:

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

pipe = pipeline("text-generation", model="tiiuae/falcon-7b-instruct", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("tiiuae/falcon-7b-instruct", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("tiiuae/falcon-7b-instruct", trust_remote_code=True)
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 tiiuae/falcon-7b-instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tiiuae/falcon-7b-instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tiiuae/falcon-7b-instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/tiiuae/falcon-7b-instruct

SGLang

How to use tiiuae/falcon-7b-instruct 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 "tiiuae/falcon-7b-instruct" \
    --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": "tiiuae/falcon-7b-instruct",
		"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 "tiiuae/falcon-7b-instruct" \
        --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": "tiiuae/falcon-7b-instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use tiiuae/falcon-7b-instruct with Docker Model Runner:
```
docker model run hf.co/tiiuae/falcon-7b-instruct
```

The correctness of the result using transformers apply_chat_template

#92

by Annorita - opened Oct 18, 2023

Discussion

Annorita

Oct 18, 2023

•

edited Oct 18, 2023

In the latest transformers (4.34.0), they have a function called "apply_chat_template" that allows us to get the prompt. For example:

from transformers import AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained("tiiuae/falcon-40b-instruct")
chat = [
  {"role": "user", "content": "USER_INSTRUCTION_1"},
  {"role": "assistant", "content": "RESPONSE_1"},
  {"role": "user", "content": "USER_INSTRUCTION_2"},
  {"role": "assistant", "content": "RESPONSE_2"},
]
res = tokenizer.apply_chat_template(chat, tokenize=False)

Falcon does not have its own tokenizer class, so transformers will directly call the PreTrainedTokenizerFast and apply the following template:
"{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}"

(check here: default_chat_template)
So the result of the example will be:

'<|im_start|>user\nUSER_INSTRUCTION_1<|im_end|>\n<|im_start|>assistant\nRESPONSE_1<|im_end|>\n<|im_start|>user\nUSER_INSTRUCTION_2<|im_end|>\n<|im_start|>assistant\nRESPONSE_2<|im_end|>\n'

However, if we encode this sentence and then decode them back, we can find that the tokenizer cannot recognize the special tokens such as <|im_start|>:

res_space = '<|im_start|>'
ids = tokenizer.encode(res_space)
tmp = []
for id in ids:
    tmp.append(tokenizer.decode(id))
#tmp = ['<', '|', 'im', '_', 'start', '|>']

Is this the right template for us to use Falcon model?

Gnurro2

Oct 25, 2023

No, that's the default template in tokenizers.
Proper format (...well, at least somewhat official): https://huggingface.co/tiiuae/falcon-7b-instruct/discussions/1#64708b0a3df93fddece002a4
Apparently the model wasn't trained on any concise format, so it seems like "whatever works". The format is the whole point of instruct training, and I really do not know why so many model trainers do not properly share the used format...

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