Instructions to use Aleph-Alpha/Pharia-1-LLM-7B-control-hf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Aleph-Alpha/Pharia-1-LLM-7B-control-hf with Transformers:

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

pipe = pipeline("text-generation", model="Aleph-Alpha/Pharia-1-LLM-7B-control-hf", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("Aleph-Alpha/Pharia-1-LLM-7B-control-hf", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Aleph-Alpha/Pharia-1-LLM-7B-control-hf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Aleph-Alpha/Pharia-1-LLM-7B-control-hf"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Aleph-Alpha/Pharia-1-LLM-7B-control-hf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Aleph-Alpha/Pharia-1-LLM-7B-control-hf

SGLang

How to use Aleph-Alpha/Pharia-1-LLM-7B-control-hf 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 "Aleph-Alpha/Pharia-1-LLM-7B-control-hf" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Aleph-Alpha/Pharia-1-LLM-7B-control-hf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "Aleph-Alpha/Pharia-1-LLM-7B-control-hf" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Aleph-Alpha/Pharia-1-LLM-7B-control-hf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Aleph-Alpha/Pharia-1-LLM-7B-control-hf with Docker Model Runner:
```
docker model run hf.co/Aleph-Alpha/Pharia-1-LLM-7B-control-hf
```

Make loss calculation possible during eval mode

by tanaymehta - opened Jan 3, 2025

base: refs/heads/main

←

from: refs/pr/7

Discussion Files changed

-1

tanaymehta

Jan 3, 2025

Previously, the loss was 0.0 during eval model which was faulty. Now the loss will be calculated during both training and eval modes.

Make loss calculation possible during eval mode68c8d4ad

tanaymehta changed pull request status to merged Jan 7, 2025

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