Instructions to use EmbeddedLLM/EAGLE-WizardLM-7B-V1.0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use EmbeddedLLM/EAGLE-WizardLM-7B-V1.0 with Transformers:

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

pipe = pipeline("text-generation", model="EmbeddedLLM/EAGLE-WizardLM-7B-V1.0")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("EmbeddedLLM/EAGLE-WizardLM-7B-V1.0")
model = AutoModelForCausalLM.from_pretrained("EmbeddedLLM/EAGLE-WizardLM-7B-V1.0")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use EmbeddedLLM/EAGLE-WizardLM-7B-V1.0 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "EmbeddedLLM/EAGLE-WizardLM-7B-V1.0"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "EmbeddedLLM/EAGLE-WizardLM-7B-V1.0",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/EmbeddedLLM/EAGLE-WizardLM-7B-V1.0

SGLang

How to use EmbeddedLLM/EAGLE-WizardLM-7B-V1.0 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 "EmbeddedLLM/EAGLE-WizardLM-7B-V1.0" \
    --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": "EmbeddedLLM/EAGLE-WizardLM-7B-V1.0",
		"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 "EmbeddedLLM/EAGLE-WizardLM-7B-V1.0" \
        --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": "EmbeddedLLM/EAGLE-WizardLM-7B-V1.0",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use EmbeddedLLM/EAGLE-WizardLM-7B-V1.0 with Docker Model Runner:
```
docker model run hf.co/EmbeddedLLM/EAGLE-WizardLM-7B-V1.0
```

tjellm commited on Jan 24, 2024

Commit

d1236ad

verified ·

1 Parent(s): 216cacc

Update README.md

Browse files

Files changed (1) hide show

README.md +11 -1

README.md CHANGED Viewed

@@ -7,4 +7,14 @@ language:
 pipeline_tag: text-generation
 tags:
 - eagle
----

 pipeline_tag: text-generation
 tags:
 - eagle
+---
+# EAGLE heads weights for WizardLM
+This model is trained using the repository https://github.com/SafeAILab/EAGLE (commit: cab744bdab6f6083fa04f19879475e5acbd1706e).
+It shows approximately 2x speed up for WizardLM model evaluate on ShareGPT dataset.
+To run the code please follow the instruction in https://github.com/SafeAILab/EAGLE .
+You will also need to download the base model weights from [WizardLM-7b-V1.0](https://huggingface.co/WizardLM/WizardLM-7B-V1.0)