Instructions to use webbrain-one/webbrain-compass-tiny-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use webbrain-one/webbrain-compass-tiny-v2 with Transformers:

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

pipe = pipeline("text-generation", model="webbrain-one/webbrain-compass-tiny-v2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("webbrain-one/webbrain-compass-tiny-v2")
model = AutoModelForCausalLM.from_pretrained("webbrain-one/webbrain-compass-tiny-v2", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use webbrain-one/webbrain-compass-tiny-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "webbrain-one/webbrain-compass-tiny-v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "webbrain-one/webbrain-compass-tiny-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/webbrain-one/webbrain-compass-tiny-v2

SGLang

How to use webbrain-one/webbrain-compass-tiny-v2 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 "webbrain-one/webbrain-compass-tiny-v2" \
    --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": "webbrain-one/webbrain-compass-tiny-v2",
		"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 "webbrain-one/webbrain-compass-tiny-v2" \
        --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": "webbrain-one/webbrain-compass-tiny-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use webbrain-one/webbrain-compass-tiny-v2 with Docker Model Runner:
```
docker model run hf.co/webbrain-one/webbrain-compass-tiny-v2
```

webbrain-compass-tiny-v2

File size: 964 Bytes

1bf438d

# Source attributions and project restrictions

Base: OpenBMB MiniCPM5-2B, revision 3497c460c89e00520c3cfa2e73f49ab7647f1177,
Apache-2.0. https://huggingface.co/openbmb/MiniCPM5-2B
Modified by WebBrain through LoRA training and full-model merging.
Original authors do not endorse this derived release.

Training dataset: https://huggingface.co/datasets/webbrain-one/webbrain-compass-v2-dataset
Preserve the source attributions, manifests and restrictions in that private repository.
WebLINX: McGill-NLP contributors, CC BY-NC-SA 4.0.
https://huggingface.co/datasets/McGill-NLP/weblinx
https://creativecommons.org/licenses/by-nc-sa/4.0/legalcode.en

Project policy restricts this derived model to noncommercial research. Preserve
all source terms, attribution and applicable share-alike obligations. Private
hosting does not grant commercial rights. This is an experimental release with
a disclosed failed behavioral gate, not a claim of general safety or success.