Instructions to use obaidtambo/urdu_ghazals_gpt2_v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use obaidtambo/urdu_ghazals_gpt2_v2 with Transformers:

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

pipe = pipeline("text-generation", model="obaidtambo/urdu_ghazals_gpt2_v2")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("obaidtambo/urdu_ghazals_gpt2_v2")
model = AutoModelForCausalLM.from_pretrained("obaidtambo/urdu_ghazals_gpt2_v2")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use obaidtambo/urdu_ghazals_gpt2_v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "obaidtambo/urdu_ghazals_gpt2_v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "obaidtambo/urdu_ghazals_gpt2_v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/obaidtambo/urdu_ghazals_gpt2_v2

SGLang

How to use obaidtambo/urdu_ghazals_gpt2_v2 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 "obaidtambo/urdu_ghazals_gpt2_v2" \
    --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": "obaidtambo/urdu_ghazals_gpt2_v2",
		"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 "obaidtambo/urdu_ghazals_gpt2_v2" \
        --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": "obaidtambo/urdu_ghazals_gpt2_v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use obaidtambo/urdu_ghazals_gpt2_v2 with Docker Model Runner:
```
docker model run hf.co/obaidtambo/urdu_ghazals_gpt2_v2
```

Urdu Ghazals Model

This model generates Urdu ghazals in response to English translated prompts.

A good prompt for this model would be an English translated line from an Urdu ghazal, such as

“kuch kehna tha aap se”.

The model will generate an Urdu ghazal in response to the prompt.

We hope you enjoy using this model to generate beautiful Urdu ghazals! The model was trained and fine-tuned on a dataset of 9000 samples of Urdu ghazals, including the famous ghazal by Mirza Ghalib:

hazāroñ ḳhvāhisheñ aisī ki har ḳhvāhish pe dam nikle

bahut nikle mire armān lekin phir bhī kam nikle

Dare kyuuñ merā qātil kyā rahegā us kī gardan par

vo ḳhuuñ jo chashm-e-tar se umr bhar yuuñ dam-ba-dam nikle

nikalnā ḳhuld se aadam kā sunte aa.e haiñ lekin

bahut be-ābrū ho kar tire kūche se ham nikle

bharam khul jaa.e zālim tere qāmat kī darāzī kā

agar is turra-e-pur-pech-o-ḳham kā pech-o-ḳham nikle

magar likhvā.e koī us ko ḳhat to ham se likhvā.e

huī sub.h aur ghar se kaan par rakh kar qalam nikle

huī is daur meñ mansūb mujh se bāda-ashāmī

phir aayā vo zamāna jo jahāñ meñ jām-e-jam nikle

huī jin se tavaqqo ḳhastagī kī daad paane kī

vo ham se bhī ziyāda ḳhasta-e-teġh-e-sitam nikle

mohabbat meñ nahīñ hai farq jiine aur marne kā

usī ko dekh kar jiite haiñ jis kāfir pe dam nikle

kahāñ mai-ḳhāne kā darvāza 'ġhālib' aur kahāñ vaa.iz

par itnā jānte haiñ kal vo jaatā thā ki ham nikle

😊

Downloads last month: 1

Safetensors

Model size

0.1B params

Tensor type

F32