Instructions to use camel-ai/CAMEL-13B-Combined-Data with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use camel-ai/CAMEL-13B-Combined-Data with Transformers:

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

pipe = pipeline("text-generation", model="camel-ai/CAMEL-13B-Combined-Data")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("camel-ai/CAMEL-13B-Combined-Data")
model = AutoModelForCausalLM.from_pretrained("camel-ai/CAMEL-13B-Combined-Data", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use camel-ai/CAMEL-13B-Combined-Data with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "camel-ai/CAMEL-13B-Combined-Data"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "camel-ai/CAMEL-13B-Combined-Data",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/camel-ai/CAMEL-13B-Combined-Data

SGLang

How to use camel-ai/CAMEL-13B-Combined-Data 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 "camel-ai/CAMEL-13B-Combined-Data" \
    --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": "camel-ai/CAMEL-13B-Combined-Data",
		"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 "camel-ai/CAMEL-13B-Combined-Data" \
        --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": "camel-ai/CAMEL-13B-Combined-Data",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use camel-ai/CAMEL-13B-Combined-Data with Docker Model Runner:
```
docker model run hf.co/camel-ai/CAMEL-13B-Combined-Data
```

Duplicate files?

by MachineThinker - opened Jun 11, 2023

Discussion

MachineThinker

Jun 11, 2023

Dear CAMEL Team, I have finished reading your paper and I am excited to test both models. However, I noticed that the model files uploaded for CAMEL-13B-Role-Playing-Data and CAMEL-13B-Combined-Data appear to be duplicates. For example, the file pytorch_model-00001-of-00006.bin has the same hash in both models (SHA256: 08d6c3bdd4c92abafdbc2e6a69d73fb9b4891cfdcd770e80bfed13673232fd04). Could you please look into this and let me know which model was actually uploaded?

itanh0b

CAMEL-AI.org org Jun 13, 2023

Hello @MachineThinker ,

We re-uploaded the models and made sure the hashes are distinct. Let us know if you're still having the same problem.

MachineThinker

Jun 13, 2023

@itanh0b Looks good. Thanks Hani.
@TheBloke You probably want to re-quantize the CAMEL-13B-Combined-Data model when you get a chance.

TheBloke

Jun 13, 2023

Oh wow! Thanks for the heads up. Huh, it's amazing it still produced coherent responses!

Did both files change? Do I need to re-do both?

TheBloke

Jun 13, 2023

Yeah I guess I do as all files changed in both repos

itanh0b

CAMEL-AI.org org Jun 13, 2023

•

edited Jun 13, 2023

@TheBloke You can requantize our models but they're obsolete now. We're working on better models. We'll release them soon. Thanks for your time and effort!

TheBloke

Jun 13, 2023

OK good to know! I just started them, may as well have them working.

But I'll look forward to the new models and quantise those as soon as they're out!

Thanks very much for the great model releases.

MachineThinker

Jun 13, 2023

@itanh0b I'm so excited to hear that new models are coming. Your efforts are much appreciated.

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