Instructions to use PhysiQuanty/Binary-LLM-POC with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use PhysiQuanty/Binary-LLM-POC with Transformers:

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

pipe = pipeline("text-generation", model="PhysiQuanty/Binary-LLM-POC", trust_remote_code=True)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("PhysiQuanty/Binary-LLM-POC", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use PhysiQuanty/Binary-LLM-POC with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/PhysiQuanty/Binary-LLM-POC

SGLang

How to use PhysiQuanty/Binary-LLM-POC 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 "PhysiQuanty/Binary-LLM-POC" \
    --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": "PhysiQuanty/Binary-LLM-POC",
		"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 "PhysiQuanty/Binary-LLM-POC" \
        --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": "PhysiQuanty/Binary-LLM-POC",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use PhysiQuanty/Binary-LLM-POC with Docker Model Runner:
```
docker model run hf.co/PhysiQuanty/Binary-LLM-POC
```

PhysiQuanty commited on May 2

Commit

1478583

verified ·

1 Parent(s): ca3d739

Update README.md

Browse files

Files changed (1) hide show

README.md +2 -1

README.md CHANGED Viewed

@@ -26,7 +26,8 @@ This repo uses `trust_remote_code=True` because it ships a custom `modeling_*.py
 (In French only, we plan 20 languages as well as scientific and mathematical knowledge for BinaryLLM1)
 - 10 millions params
-- 2 billions pre-training tokens (without SFT)
 - 40k steps
 - 1e-4 learning rate
 - Fp32 weight, FSDP training on 8 NVIDIA V100

 (In French only, we plan 20 languages as well as scientific and mathematical knowledge for BinaryLLM1)
 - 10 millions params
+- 2 billions pre-training tokens
+- Without SFT, pre-train only
 - 40k steps
 - 1e-4 learning rate
 - Fp32 weight, FSDP training on 8 NVIDIA V100