Instructions to use marathi-llm/MahaMarathi-7B-v24.01-Base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use marathi-llm/MahaMarathi-7B-v24.01-Base with Transformers:

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

pipe = pipeline("text-generation", model="marathi-llm/MahaMarathi-7B-v24.01-Base")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("marathi-llm/MahaMarathi-7B-v24.01-Base")
model = AutoModelForCausalLM.from_pretrained("marathi-llm/MahaMarathi-7B-v24.01-Base")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use marathi-llm/MahaMarathi-7B-v24.01-Base with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "marathi-llm/MahaMarathi-7B-v24.01-Base"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "marathi-llm/MahaMarathi-7B-v24.01-Base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/marathi-llm/MahaMarathi-7B-v24.01-Base

SGLang

How to use marathi-llm/MahaMarathi-7B-v24.01-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 "marathi-llm/MahaMarathi-7B-v24.01-Base" \
    --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": "marathi-llm/MahaMarathi-7B-v24.01-Base",
		"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 "marathi-llm/MahaMarathi-7B-v24.01-Base" \
        --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": "marathi-llm/MahaMarathi-7B-v24.01-Base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use marathi-llm/MahaMarathi-7B-v24.01-Base with Docker Model Runner:
```
docker model run hf.co/marathi-llm/MahaMarathi-7B-v24.01-Base
```

You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

MahaMarathi 7B, is a domain adapted, continually pre-trained, and instruction fine-tuned native Marathi large language model (LLM) with 7 billion parameters based on Llama2+Mistral, and trained on a large corpus of Marathi text.
महामराठी हे 7अब्ज पॅरामीटर्स वापरून पूर्व-प्रशिक्षित आणि सुचनांच्या आधारे फाईन ट्यून केलेले मूळ मराठी महाभाषा समीकरण संच आहे ज्याला लार्ज लँग्वेज मॉडेल (एल. एल. एम.) असेही म्हणतात. 83 दशलक्षाहून अधिक स्थानिक भाषिकांसाठी तयार केलेले हे कृत्रिम बुद्धिमत्तेवर (ए. आय.) आधारित तंत्रज्ञान आहे जे मराठीतील कठीण संभाषणे आणि सूचना सहज हाताळण्यास सक्षम आहे.
This released model is a base model and not meant to be used as is. It is recommended to first finetune it on downstream tasks that you are interested in.

# Usage
import torch
from transformers import LlamaTokenizer, LlamaForCausalLM

tokenizer = LlamaTokenizer.from_pretrained('marathi-llm/MahaMarathi-7B-v24.01-Base')
model = LlamaForCausalLM.from_pretrained('marathi-llm/MahaMarathi-7B-v24.01-Base', torch_dtype=torch.bfloat16)

prompt = "मी एक ए. आय. द्वारा तयार केलेले महाभाषा समीकरण संच आहे."
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]

Example 1	Example 2

Downloads last month: 40

Safetensors

Model size

7B params

Tensor type

F16

marathi-llm
/

MahaMarathi-7B-v24.01-Base

You need to agree to share your contact information to access this model

Spaces using marathi-llm/MahaMarathi-7B-v24.01-Base 4