Instructions to use URajinda/ShweYon-V3-Base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use URajinda/ShweYon-V3-Base with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="URajinda/ShweYon-V3-Base") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("URajinda/ShweYon-V3-Base") model = AutoModelForCausalLM.from_pretrained("URajinda/ShweYon-V3-Base", 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=256) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use URajinda/ShweYon-V3-Base with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "URajinda/ShweYon-V3-Base" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "URajinda/ShweYon-V3-Base", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/URajinda/ShweYon-V3-Base
- SGLang
How to use URajinda/ShweYon-V3-Base 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 "URajinda/ShweYon-V3-Base" \ --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": "URajinda/ShweYon-V3-Base", "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 "URajinda/ShweYon-V3-Base" \ --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": "URajinda/ShweYon-V3-Base", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use URajinda/ShweYon-V3-Base with Docker Model Runner:
docker model run hf.co/URajinda/ShweYon-V3-Base
đ° ShweYon-V3-Base (áá˝ážáąááŻááş-V3)
ShweYon-V3-Base áááş Qwen 2.5 1.5B ááᯠáĄááźáąááśá ááźááşááŹááŹááŹá ááŹá¸áĄáá˝ááş áĄáá°á¸ááźáŻááźááşááŹá¸ááąáŹ Base Model ááźá áşááŤáááşá ᤠVersion áá˝ááş ááááş Version ááťáŹá¸áá˛áˇáááŻáˇ Tokenizer ááŽá¸ááźáŹá¸ááŻáśá¸áááş ááááŻááąáŹáˇáᲠModel á Embedding áá˛áááŻáˇ ááźááşááŹááŻáśáááşááťáŹá¸ááᯠáááŻááşáááŻááşááąáŤááşá¸á ááşááŹá¸ááŤáááşá ShweYon-V3-Base is a Myanmar-centric base language model built on top of the Qwen 2.5 1.5B architecture. This model is a milestone in the "ShweYon" project, focusing on improving the efficiency of Myanmar script processing through a custom tokenizer.
đŻ Purpose (áááşáá˝ááşááťááş)
ᤠModel áááş ááźááşááŹááŹááŹá ááŹá¸áĄáá˝ááş Foundation Base Model áá áşááŻáĄááźá áş áááşáá˝ááşááŤáááşá ᤠModel ááᯠáĄááźáąááśá Chatbot ááťáŹá¸á Question Answering á áá áşááťáŹá¸áážááˇáş áĄááźáŹá¸ááąáŹ Downstream NLP Task ááťáŹá¸áĄáá˝ááş áááşááśá Fine-tuning (SFT/RLHF) ááźáŻááŻááşáááş áĄááąáŹááşá¸ááŻáśá¸ áĄáŻááşááźá áşááźá áşááŤáááşá
⨠Technical Highlights
- Integrated Tokenizer: ááźááşááŹáááááşááťáŹá¸áážááˇáş á ááŹá¸ááŻáśá¸ááąáŤááşá¸ á,ááá ááťáąáŹáş ááŤáááşááąáŹ Custom Tokenizer ááᯠáá áşááŤáááşá¸ áááˇáşáá˝ááşá¸ááŹá¸ááŤáááşá
- Extended Vocabulary: Vocabulary Size áááŻ
160,746áĄáá áááŻá¸ááźážááˇáşááŹá¸áááźááˇáş ááźááşááŹá áŹááŹá¸ááťáŹá¸ááᯠáááŻáááŻááťá áşááťá áşá á˝áŹáážááˇáş ááźááşáááşá á˝áŹ áá˝ááşááťááşáááŻááşááŤáááşá - Base Training: ááźááşááŹá áŹááą á áŹáĄáŻááşááťáŹá¸á á˝áŹááźááˇáş Model á ááźááşááŹá ᏠáĄááźáąááśáááŻááŻá áááŻáááŻááąáŹááşá¸áá˝ááşááŹá áąáááş ááąáˇááťááˇáşááąá¸ááŹá¸ááŤáááşá
đ Quick Start
ᤠBase Model ááᯠáĄáąáŹááşááŤáĄáááŻááşá¸ ááąáŤáşáá°áĄááŻáśá¸ááźáŻáááŻááşááŤáááşá
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "URajinda/ShweYon-V3-Base"
# ááťááşáážááşááąáŹ áŚá¸áážáąáŹááşááąáŹ áá
áşááŤáááşá¸ ááŤááŹááŤáááş
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)
# á
ááşá¸áááşááźááˇáşáááş
prompt = "ááźááşááŹáááŻááşááśá ááááŻááşá¸ááźáąáŹááşá¸áážáŹ"
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=32)
print(tokenizer.decode(outputs[0]))
â ď¸ Note
ᤠModel áááş Base Model áᏠááźá
áşááąáŹááźáąáŹááˇáş áá°ááŹá¸áážááˇáş á
ááŹá¸ááźáąáŹáááŻáááş (Instruction Following) áĄáá˝ááş áááşááśá Chat Fine-tuning ááŻááşáááş áááŻáĄááşááŤááąá¸áááşá
âď¸ License
Apache License 2.0
- Downloads last month
- 13