Instructions to use NucleusAI/RetNet-410m-XATL with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use NucleusAI/RetNet-410m-XATL with Transformers:

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

pipe = pipeline("text-generation", model="NucleusAI/RetNet-410m-XATL", trust_remote_code=True)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("NucleusAI/RetNet-410m-XATL", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use NucleusAI/RetNet-410m-XATL with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "NucleusAI/RetNet-410m-XATL"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NucleusAI/RetNet-410m-XATL",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/NucleusAI/RetNet-410m-XATL

SGLang

How to use NucleusAI/RetNet-410m-XATL 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 "NucleusAI/RetNet-410m-XATL" \
    --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": "NucleusAI/RetNet-410m-XATL",
		"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 "NucleusAI/RetNet-410m-XATL" \
        --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": "NucleusAI/RetNet-410m-XATL",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use NucleusAI/RetNet-410m-XATL with Docker Model Runner:
```
docker model run hf.co/NucleusAI/RetNet-410m-XATL
```

syncdoth commited on Jul 28, 2024

Commit

ac26f6f

verified ·

1 Parent(s): af3b01e

Update README.md

Browse files

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -10,7 +10,7 @@ language:
 This is a [RetNet](https://arxiv.org/abs/2307.08621) model, accompanying the paper [Cross-Architecture Transfer Learning for Linear-Cost Inference Transformers](https://arxiv.org/abs/2404.02684v1),
 In this work, we proposed to *not* train new Linear-Cost Inference models (e.g. RetNet) from scratch, but to transfer shared weight components from other PTLMs.
-The model's input/output embeddings, MLP weights, Layer Norms, Attention Output Projections ($W_O$) has been transferred from [pythia-1B](https://huggingface.co/EleutherAI/pythia-410m). For more detail, please refer to the paper.
 ## Model Details

 This is a [RetNet](https://arxiv.org/abs/2307.08621) model, accompanying the paper [Cross-Architecture Transfer Learning for Linear-Cost Inference Transformers](https://arxiv.org/abs/2404.02684v1),
 In this work, we proposed to *not* train new Linear-Cost Inference models (e.g. RetNet) from scratch, but to transfer shared weight components from other PTLMs.
+The model's input/output embeddings, MLP weights, Layer Norms, Attention Output Projections ($W_O$) has been transferred from [pythia-410m](https://huggingface.co/EleutherAI/pythia-410m). For more detail, please refer to the paper.
 ## Model Details