Instructions to use CohereLabs/c4ai-command-r-v01-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use CohereLabs/c4ai-command-r-v01-4bit with Transformers:

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

pipe = pipeline("text-generation", model="CohereLabs/c4ai-command-r-v01-4bit", 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("CohereLabs/c4ai-command-r-v01-4bit", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("CohereLabs/c4ai-command-r-v01-4bit", 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 CohereLabs/c4ai-command-r-v01-4bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "CohereLabs/c4ai-command-r-v01-4bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CohereLabs/c4ai-command-r-v01-4bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/CohereLabs/c4ai-command-r-v01-4bit

SGLang

How to use CohereLabs/c4ai-command-r-v01-4bit 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 "CohereLabs/c4ai-command-r-v01-4bit" \
    --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": "CohereLabs/c4ai-command-r-v01-4bit",
		"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 "CohereLabs/c4ai-command-r-v01-4bit" \
        --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": "CohereLabs/c4ai-command-r-v01-4bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use CohereLabs/c4ai-command-r-v01-4bit with Docker Model Runner:
```
docker model run hf.co/CohereLabs/c4ai-command-r-v01-4bit
```

OSError: CohereForAI/c4ai-command-r-v01 does not appear to have a file named configuration_cohere.py

by ahkarami - opened Apr 14, 2024

Discussion

ahkarami

Apr 14, 2024

Hi,
When I want to use the model, I got following error:

OSError: CohereForAI/c4ai-command-r-v01 does not appear to have a file named configuration_cohere.py. Checkout 'https://huggingface.co/CohereForAI/c4ai-command-r-v01/main' for available files.

Would you please help me to address this issue?

drinkingxi

Apr 16, 2024

update transformers to 4.39.1 or higher

ahkarami

Apr 17, 2024

•

edited Apr 17, 2024

@drinkingxi
Thanks. I have used transformers==4.39.3 and python=3.9.16, but I have faced above error. Would you please help me to address it? (In fact, I downloaded the model, about 20 days ago, and then move it from my home/.cache/huggingface/hub to another folder and then when I want to load it from this new address I faced above error. When I remove it completely and want to download it again, I also faced above error.)

leonaci

Apr 24, 2024

same error.
I resolved it by removing the auto_map key from config.json and adding exist_ok=True to the arguments of the register function in the last lines of configuration_cohere.py and tokenization_cohere_fast.py.
I'm not sure why this is happening, but...

ahkarami

Apr 26, 2024

@leonaci
Thanks a lot for your help. But, I got following error again (after removing the auto_map key from config.json ):

OSError: It looks like the config file at '/HuggingFace_Models/c4ai/models--CohereForAI--c4ai-command-r-v01-4bit/snapshots/183f6e1182a46c223dde00ebf1ace9194ea221aa/config.json' is not a valid JSON file.

leonaci

May 2, 2024

@ahkarami

not a valid JSON file

Are you sure there aren't any syntax errors in your JSON? (like a trailing comma or something...)
Check it out carefully again.

epignatelli

May 19, 2024

Why is such a fundamental issue still open after a month?

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