Openhathi 7B Hi V0.1 Base

🚀 OpenHathi-7B-Hi-v0.1-Base

The first model in the OpenHathi series by Sarvam AI, a 7B parameter model based on Llama2, trained on Hindi, English, and Hinglish.

🚀 Quick Start

This repository hosts the first model in the OpenHathi series of models to be released by Sarvam AI. It is a 7B parameter model, built upon Llama2, and trained on Hindi, English, and Hinglish. More details about the model, its training procedure, and evaluations can be found here.

⚠️ Important Note

This is a base model and not meant to be used as is. We recommend first finetuning it on task(s) you are interested in.

💻 Usage Examples

Basic Usage

# Usage
import torch
from transformers import LlamaTokenizer, LlamaForCausalLM

tokenizer = LlamaTokenizer.from_pretrained('sarvamai/OpenHathi-7B-Hi-v0.1-Base')
model = LlamaForCausalLM.from_pretrained('sarvamai/OpenHathi-7B-Hi-v0.1-Base', torch_dtype=torch.bfloat16)

prompt = "मैं एक अच्छा हाथी हूँ"
inputs = tokenizer(prompt, return_tensors="pt")

# Generate
generate_ids = model.generate(inputs.input_ids, max_length=30)
tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]

📄 License

This model is under the Llama2 license.