Instructions to use Rapnss/VIA-01 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Rapnss/VIA-01 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="Rapnss/VIA-01") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Rapnss/VIA-01", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use Rapnss/VIA-01 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Rapnss/VIA-01" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Rapnss/VIA-01", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/Rapnss/VIA-01
- SGLang
How to use Rapnss/VIA-01 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 "Rapnss/VIA-01" \ --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": "Rapnss/VIA-01", "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 "Rapnss/VIA-01" \ --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": "Rapnss/VIA-01", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use Rapnss/VIA-01 with Docker Model Runner:
docker model run hf.co/Rapnss/VIA-01
First Indian text-generation llm model with 7.24B parameters
VIA-01 by Rapnss VIA-01 is a state-of-the-art language model crafted by Rapnss, designed for exceptional performance in conversational tasks and code generation. Engineered for speed and versatility, VIA-01 delivers precise, high-quality responses across diverse applications, from answering complex questions to generating efficient code snippets. Features
Dual-Purpose Excellence: Seamlessly blends natural language understanding with robust code generation, ideal for developers and conversational AI use cases. Optimized Performance: Lightweight design ensures fast inference, even in resource-constrained environments. Easy Integration: Compatible with Hugging Face Transformers for straightforward deployment. Rapnss Innovation: A unique AI built to inspire creativity and productivity.
Usage Get started with VIA-01 using the following Python code:
from transformers import pipeline
import torch
# Initialize the pipeline
pipe = pipeline(
"text-generation",
model="rapnss/VIA-01",
torch_dtype=torch.float16,
device_map="auto",
max_new_tokens=15
)
# Generate a response
prompt = "Write a Python function to sort a list:"
response = pipe(prompt)[0]['generated_text']
print(response)
Example Output:
Write a Python function to sort a list:
def sort_list(arr):
return sorted(arr)
## Installation
Install required dependencies:
```bash
pip install transformers torch accelerate gradio
Performance
Inference Speed: Optimized for low-latency responses, typically ~20-40 seconds on standard CPU hardware (e.g., Hugging Face free Space). For sub-10-second responses, use a GPU-enabled environment (e.g., Hugging Face Pro Space). Model Size: ~8GB, balanced for efficiency and performance. Capabilities: Excels in conversational queries, technical problem-solving, and code generation tasks like writing functions or debugging snippets.
Try It Out
Interact with VIA-01 via our Hugging Face Space, featuring a Gradio interface for real-time testing. Limitations
Response Length: Short responses (up to 15 tokens) recommended for optimal speed on free-tier hosting. Hardware: CPU-based inference may be slower than GPU; performance varies with deployment setup.
License
Licensed under the Apache 2.0 License, enabling flexible use and redistribution. Contact Created by Rapnss. For inquiries or feedback, reach out via Hugging Face or the VIA-01 Space.
Made with ❤️ by Rapnss.
- Downloads last month
- 4