mideind/icelandic-common-crawl-corpus-IC3
Viewer • Updated • 1.97M • 107 • 1
How to use Sigurdur/ice-gpt-cc with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("text-generation", model="Sigurdur/ice-gpt-cc") # Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("Sigurdur/ice-gpt-cc")
model = AutoModelForCausalLM.from_pretrained("Sigurdur/ice-gpt-cc")How to use Sigurdur/ice-gpt-cc with vLLM:
# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Sigurdur/ice-gpt-cc"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "Sigurdur/ice-gpt-cc",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'docker model run hf.co/Sigurdur/ice-gpt-cc
How to use Sigurdur/ice-gpt-cc with SGLang:
# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
--model-path "Sigurdur/ice-gpt-cc" \
--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": "Sigurdur/ice-gpt-cc",
"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 "Sigurdur/ice-gpt-cc" \
--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": "Sigurdur/ice-gpt-cc",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'How to use Sigurdur/ice-gpt-cc with Docker Model Runner:
docker model run hf.co/Sigurdur/ice-gpt-cc
A light-wheight gpt model trained on the icelandic-common-crawl-corpus from Mideind. It uses ice-tokenizer a bpe tokenizer for both training and inference.
Sigurdur Haukur Birgisson