Instructions to use akoksal/LongForm-OPT-6.7B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use akoksal/LongForm-OPT-6.7B with Transformers:

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

pipe = pipeline("text-generation", model="akoksal/LongForm-OPT-6.7B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("akoksal/LongForm-OPT-6.7B")
model = AutoModelForCausalLM.from_pretrained("akoksal/LongForm-OPT-6.7B")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use akoksal/LongForm-OPT-6.7B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "akoksal/LongForm-OPT-6.7B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "akoksal/LongForm-OPT-6.7B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/akoksal/LongForm-OPT-6.7B

SGLang

How to use akoksal/LongForm-OPT-6.7B 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 "akoksal/LongForm-OPT-6.7B" \
    --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": "akoksal/LongForm-OPT-6.7B",
		"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 "akoksal/LongForm-OPT-6.7B" \
        --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": "akoksal/LongForm-OPT-6.7B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use akoksal/LongForm-OPT-6.7B with Docker Model Runner:
```
docker model run hf.co/akoksal/LongForm-OPT-6.7B
```

Why is this so slow?

by adivekar - opened May 28, 2023

Discussion

adivekar

May 28, 2023

It takes 3+ minutes to generate a single text (using max_new_tokens=512, input max_length=1024).

Is this expected?

akoksal

Owner May 31, 2023

Yes, unfortunately it is expected. This model tends to create longer pieces of texts, which naturally requires more time for generation (until it sees the eos token). If you reduce the value of max_new_tokens to something like 50, the generation would be significantly faster but the output probably will be truncated.

akoksal changed discussion status to closed May 31, 2023

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