Instructions to use openai-community/gpt2-large with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use openai-community/gpt2-large with Transformers:

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

pipe = pipeline("text-generation", model="openai-community/gpt2-large")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("openai-community/gpt2-large")
model = AutoModelForCausalLM.from_pretrained("openai-community/gpt2-large")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use openai-community/gpt2-large with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "openai-community/gpt2-large"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "openai-community/gpt2-large",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/openai-community/gpt2-large

SGLang

How to use openai-community/gpt2-large 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 "openai-community/gpt2-large" \
    --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": "openai-community/gpt2-large",
		"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 "openai-community/gpt2-large" \
        --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": "openai-community/gpt2-large",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use openai-community/gpt2-large with Docker Model Runner:
```
docker model run hf.co/openai-community/gpt2-large
```

No module named 'keras.saving.legacy'

by fccoelho - opened Jan 20, 2023

Discussion

fccoelho

Jan 20, 2023

Getting this exception when trying to run the examples from the documentation

lysandre

OpenAI community org Jan 20, 2023

Hey @fccoelho , could you share which example you're trying to run?

cc @Rocketknight1 @joaogante

Rocketknight1

Jan 20, 2023

Hi @fccoelho , TF engineer here! Can you also share the version of TensorFlow and the version of Transformers you're using? We'll try to reproduce the problem here.

fccoelho

Jan 20, 2023

Transformers: 4.25.1
tensorflow: 2.11.0

fccoelho

Jan 21, 2023

@lysandre

from transformers import pipeline, set_seed
generator = pipeline('text-generation', model='gpt2-large')
set_seed(42)
generator("Hello, I'm a language model,", max_length=30, num_return_sequences=5)

Rocketknight1

Jan 23, 2023

Hi @fccoelho - we think this might be an incompatibility with TF 2.11 and that version of Transformers. We've added some patches to enable 2.11 support in Transformers but they haven't been released yet. Can you try installing transformers from main with the following code, and seeing if that resolves the issue? pip install --upgrade git+https://github.com/huggingface/transformers.git

fccoelho

Jan 25, 2023

Thanks @Rocketknight1 , I tried again, after installing the version from main, but I am still getting the same Exception.

mosjel63

May 9, 2024

I tried the same example as we are now in 2024 and still the same error.
Transformers version: 4.28.1
Tensorflow version: 2.16.1

joaogante

May 14, 2024

@mosjel63 👋

You'll need to use an older TF version or a newer transformers version :)

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