Instructions to use myrkur/paya with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use myrkur/paya with Transformers:

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

pipe = pipeline("text-generation", model="myrkur/paya")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("myrkur/paya")
model = AutoModelForCausalLM.from_pretrained("myrkur/paya")
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

vLLM

How to use myrkur/paya with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/myrkur/paya

SGLang

How to use myrkur/paya 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 "myrkur/paya" \
    --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": "myrkur/paya",
		"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 "myrkur/paya" \
        --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": "myrkur/paya",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use myrkur/paya with Docker Model Runner:
```
docker model run hf.co/myrkur/paya
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Paya (aya 23 8B Instruction Tuned on Farsi)

Welcome to PAYA, a powerful Persian text generation model built upon the foundations of Aya 23 8B, a multilingual language model. PAYA has been fine-tuned using the supervised finetuning technique, employing the DORA method for efficient refinement on Persian datasets, particularly leveraging the persian-alpaca-deep-clean dataset.

Features

Advanced Text Generation: Generate coherent and contextually relevant Persian text with ease.
Efficient Fine-Tuning: Utilizes the DORA method for streamlined fine-tuning on Persian datasets.
Optimized Tokenization: The model's tokenizer ensures accurate representation of Persian words, enhancing the quality of generated text.

Usage

You can quickly get started with PAYA using the following sample code:

import transformers
import torch

model_id = "myrkur/paya"

pipeline = transformers.pipeline(
    "text-generation",
    model=model_id,
    model_kwargs={"torch_dtype": torch.bfloat16},
    device_map="auto",
)

messages = [
    {"role": "user", "content": "علم بهتر است یا ثروت؟"},
]

prompt = pipeline.tokenizer.apply_chat_template(
        messages, 
        tokenize=False, 
        add_generation_prompt=True
)

terminators = [
    pipeline.tokenizer.eos_token_id,
]

outputs = pipeline(
    prompt,
    max_new_tokens=512,
    eos_token_id=terminators,
    do_sample=True,
    temperature=0.4,
    top_p=0.9,
    repetition_penalty=1.1
)
print(outputs[0]["generated_text"][len(prompt):])

Why PAYA?

PAYA stands out for its exceptional tokenization capabilities, accurately capturing the nuances of the Persian language. Additionally, its fine-tuned parameters and efficient training methodology ensure remarkable results in text generation tasks.

Contributions

Contributions to PAYA are welcome! Whether it's enhancing the model's capabilities, improving its performance on specific tasks, or evaluating its performance, your contributions can help advance Persian natural language processing.