Instructions to use madatnlp/torch-trinity with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use madatnlp/torch-trinity with Transformers:

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

pipe = pipeline("text-generation", model="madatnlp/torch-trinity")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("madatnlp/torch-trinity")
model = AutoModelForCausalLM.from_pretrained("madatnlp/torch-trinity")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use madatnlp/torch-trinity with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/madatnlp/torch-trinity

SGLang

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

Docker Model Runner
How to use madatnlp/torch-trinity with Docker Model Runner:
```
docker model run hf.co/madatnlp/torch-trinity
```

madatnlp
/

torch-trinity

add tokenizer

+version https://git-lfs.github.com/spec/v1
+oid sha256:cdedf0d487fc4d2977408613b5e096363a1d133bace8b6b296f8fb5afa4b2470
+size 5161

+version https://git-lfs.github.com/spec/v1
+oid sha256:49fd0763ad74257d144b9fbcaf217c6b2b4dd1316f3babbfd43b471b242161ea
+size 15151

+version https://git-lfs.github.com/spec/v1
+oid sha256:3fb935c5b5177d003602540324d91d2199490245e93e5a632a332b59eea94437
+size 21300

+version https://git-lfs.github.com/spec/v1
+oid sha256:ed93866db493228c5c565d655462ec48493649f2d3f1ee013fddd2cb99fe6543
+size 6137


1	+ {"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": "<mask>"}


1	+ {"unk_token": "<\|endoftext\|>", "bos_token": "<s>", "eos_token": "</s>", "add_prefix_space": true, "special_tokens_map_file": "/home/tov324/.cache/huggingface/transformers/c2ab65b9d700d0871fd407d489869d7b93f69fb5f1a58fb1fac796fd43b9ea27.1f5b09bb43973b9fbd2ba75c9fe44ffab036b980c4e6a9d779aa7707913416fe", "name_or_path": "skt/ko-gpt-trinity-1.2B-v0.5", "tokenizer_class": "GPT2Tokenizer"}

	@@ -0,0 +1 @@


1	+ checkpoint-*/

	@@ -0,0 +1,3 @@

	@@ -0,0 +1,3 @@