Instructions to use jjjlimaus/sn38-r7-2016-winner with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use jjjlimaus/sn38-r7-2016-winner with Transformers:

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

pipe = pipeline("text-generation", model="jjjlimaus/sn38-r7-2016-winner")

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("jjjlimaus/sn38-r7-2016-winner", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use jjjlimaus/sn38-r7-2016-winner with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "jjjlimaus/sn38-r7-2016-winner"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "jjjlimaus/sn38-r7-2016-winner",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/jjjlimaus/sn38-r7-2016-winner

SGLang

How to use jjjlimaus/sn38-r7-2016-winner 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 "jjjlimaus/sn38-r7-2016-winner" \
    --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": "jjjlimaus/sn38-r7-2016-winner",
		"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 "jjjlimaus/sn38-r7-2016-winner" \
        --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": "jjjlimaus/sn38-r7-2016-winner",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use jjjlimaus/sn38-r7-2016-winner with Docker Model Runner:
```
docker model run hf.co/jjjlimaus/sn38-r7-2016-winner
```

You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

sn38-r7-2016-winner

Rehost of the round 7 (cutoff 2016) SN38 winner.


Round	7
Cutoff year	2016
UID	168
Architecture	Nanoexpand
Leak score	-19.1233
Quality win rate	1.000
Final	1.000
Original	`beatrizh/allpass-1@d75f42facdc18eae1a8e7af826f1b60d91b872ba`

Weights are copied from the original public Hugging Face pin. This repo is public and gated (manual) so downloads require an approved HF account.

Not an original training run. Credit belongs to the submitting miner / original repo.

Downloads last month: -

Safetensors

Model size

2B params

Tensor type

BF16