allenai/c4
Viewer • Updated • 10.4B • 780k • 586
How to use domce20/GPT2-Lithuanian with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("text-generation", model="domce20/GPT2-Lithuanian") # Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("domce20/GPT2-Lithuanian")
model = AutoModelForCausalLM.from_pretrained("domce20/GPT2-Lithuanian")How to use domce20/GPT2-Lithuanian with vLLM:
# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "domce20/GPT2-Lithuanian"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "domce20/GPT2-Lithuanian",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'docker model run hf.co/domce20/GPT2-Lithuanian
How to use domce20/GPT2-Lithuanian with SGLang:
# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
--model-path "domce20/GPT2-Lithuanian" \
--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": "domce20/GPT2-Lithuanian",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'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 "domce20/GPT2-Lithuanian" \
--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": "domce20/GPT2-Lithuanian",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'How to use domce20/GPT2-Lithuanian with Docker Model Runner:
docker model run hf.co/domce20/GPT2-Lithuanian
GPT-2 based model trained for Lithuanian.
The model architecture is copied from the ai-forever/mGPT model, however it is trained from scratch on a modified partition of the Lithuanian partition of the mC4 dataset.
The training was done on Vilnius University supercomputer.