Text Generation
Transformers
Safetensors
PyTorch
English
wiola
decoder-only
causal-language-model
research
custom_code
Instructions to use oscowlai/Wiola360M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use oscowlai/Wiola360M with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="oscowlai/Wiola360M", trust_remote_code=True)# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("oscowlai/Wiola360M", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use oscowlai/Wiola360M with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "oscowlai/Wiola360M" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "oscowlai/Wiola360M", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/oscowlai/Wiola360M
- SGLang
How to use oscowlai/Wiola360M 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 "oscowlai/Wiola360M" \ --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": "oscowlai/Wiola360M", "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 "oscowlai/Wiola360M" \ --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": "oscowlai/Wiola360M", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use oscowlai/Wiola360M with Docker Model Runner:
docker model run hf.co/oscowlai/Wiola360M
| # coding=utf-8 | |
| """Wiola: a small language model with SRPE, GCLA, ATM, DSFF and WiolaRMSNorm. | |
| Importing this package registers Wiola with the HuggingFace Auto* classes so | |
| that ``AutoModelForCausalLM.from_pretrained("oscowlai/wiola-360m")`` works once | |
| weights are published. | |
| """ | |
| from .configuration_wiola import WiolaConfig | |
| from .modeling_wiola import ( | |
| WiolaDecoderLayer, | |
| WiolaForCausalLM, | |
| WiolaModel, | |
| WiolaPreTrainedModel, | |
| ) | |
| __version__ = "0.1.1" | |
| __all__ = [ | |
| "WiolaConfig", | |
| "WiolaModel", | |
| "WiolaForCausalLM", | |
| "WiolaPreTrainedModel", | |
| "WiolaDecoderLayer", | |
| ] | |
| def _register_auto_classes(): | |
| try: | |
| from transformers import AutoConfig, AutoModel, AutoModelForCausalLM | |
| except Exception: # transformers not installed | |
| return | |
| try: | |
| AutoConfig.register("wiola", WiolaConfig) | |
| AutoModel.register(WiolaConfig, WiolaModel) | |
| AutoModelForCausalLM.register(WiolaConfig, WiolaForCausalLM) | |
| except Exception: | |
| # Already registered (e.g. re-import) — safe to ignore. | |
| pass | |
| # The auto_map in config.json now handles Hub loading; we no longer call | |
| # register_for_auto_class to avoid save_pretrained() packaging source code. | |
| # WiolaConfig.register_for_auto_class("AutoConfig") | |
| # WiolaModel.register_for_auto_class("AutoModel") | |
| # WiolaForCausalLM.register_for_auto_class("AutoModelForCausalLM") | |
| _register_auto_classes() |