Instructions to use DedeProGames/GPT-U-20M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use DedeProGames/GPT-U-20M with Transformers:

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

pipe = pipeline("text-generation", model="DedeProGames/GPT-U-20M")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("DedeProGames/GPT-U-20M")
model = AutoModelForCausalLM.from_pretrained("DedeProGames/GPT-U-20M", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use DedeProGames/GPT-U-20M with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "DedeProGames/GPT-U-20M"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DedeProGames/GPT-U-20M",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/DedeProGames/GPT-U-20M

SGLang

How to use DedeProGames/GPT-U-20M 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 "DedeProGames/GPT-U-20M" \
    --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": "DedeProGames/GPT-U-20M",
		"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 "DedeProGames/GPT-U-20M" \
        --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": "DedeProGames/GPT-U-20M",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use DedeProGames/GPT-U-20M with Docker Model Runner:
```
docker model run hf.co/DedeProGames/GPT-U-20M
```

GPT-U-20M / scripts /run_train.sh

DedeProGames

GPT-U-20M: final weights, tokenizer, model card, scripts and logs

13caab9 verified 24 days ago

Raw History Blame Contribute Delete

1 kB

#!/usr/bin/env bash
# Start GPT-U-20M training in a detached tmux session with a persistent log. Re-running resumes automatically.
#   scripts/run_train.sh            -> session "gptu",  full run, log logs/gptu.out
#   SESSION=smoke scripts/run_train.sh --smoke
set -euo pipefail
cd "$(dirname "$0")/.."
SESSION=${SESSION:-gptu}
VENV=${VENV:-$HOME/venvs/gpt-u}
CACHE=${CACHE:-/mnt/c/Users/andre/gpt-u-20m-data/cache}  # compiler caches off the exFAT-hosted WSL disk
ENVS="NCCL_P2P_DISABLE=1 NCCL_IB_DISABLE=1 CUDA_DEVICE_ORDER=PCI_BUS_ID PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True OMP_NUM_THREADS=2 TORCHINDUCTOR_CACHE_DIR=$CACHE/inductor TRITON_CACHE_DIR=$CACHE/triton"
# The env goes inside the command: a running tmux server would not inherit this shell's exports.
tmux new-session -d -s "$SESSION" \
  "cd $PWD && env $ENVS $VENV/bin/torchrun --standalone --nproc_per_node=2 scripts/train.py $* 2>&1 | tee -a logs/$SESSION.out"
echo "tmux session '$SESSION' started, log: logs/$SESSION.out"

	#!/usr/bin/env bash
	# Start GPT-U-20M training in a detached tmux session with a persistent log. Re-running resumes automatically.
	# scripts/run_train.sh -> session "gptu", full run, log logs/gptu.out
	# SESSION=smoke scripts/run_train.sh --smoke
	set -euo pipefail
	cd "$(dirname "$0")/.."
	SESSION=${SESSION:-gptu}
	VENV=${VENV:-$HOME/venvs/gpt-u}
	CACHE=${CACHE:-/mnt/c/Users/andre/gpt-u-20m-data/cache} # compiler caches off the exFAT-hosted WSL disk
	ENVS="NCCL_P2P_DISABLE=1 NCCL_IB_DISABLE=1 CUDA_DEVICE_ORDER=PCI_BUS_ID PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True OMP_NUM_THREADS=2 TORCHINDUCTOR_CACHE_DIR=$CACHE/inductor TRITON_CACHE_DIR=$CACHE/triton"
	# The env goes inside the command: a running tmux server would not inherit this shell's exports.
	tmux new-session -d -s "$SESSION" \
	"cd $PWD && env $ENVS $VENV/bin/torchrun --standalone --nproc_per_node=2 scripts/train.py $* 2>&1 \| tee -a logs/$SESSION.out"
	echo "tmux session '$SESSION' started, log: logs/$SESSION.out"