Instructions to use mecoffey/NPC_brain with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries
PEFT
How to use mecoffey/NPC_brain with PEFT:
```
Task type is invalid.
```

How to use mecoffey/NPC_brain with Transformers:

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

pipe = pipeline("text-generation", model="mecoffey/NPC_brain")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mecoffey/NPC_brain")
model = AutoModelForCausalLM.from_pretrained("mecoffey/NPC_brain")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use mecoffey/NPC_brain with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mecoffey/NPC_brain"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mecoffey/NPC_brain",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/mecoffey/NPC_brain

SGLang

How to use mecoffey/NPC_brain 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 "mecoffey/NPC_brain" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mecoffey/NPC_brain",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "mecoffey/NPC_brain" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mecoffey/NPC_brain",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use mecoffey/NPC_brain with Docker Model Runner:
```
docker model run hf.co/mecoffey/NPC_brain
```

NPC_brain / README.md

mecoffey

finalized param count

a3f8e5f 17 days ago

preview code

Raw

History Blame Contribute Delete

1.31 kB

	---
	base_model: openbmb/MiniCPM5-1B
	library_name: peft
	pipeline_tag: text-generation
	tags:
	- base_model:adapter:openbmb/MiniCPM5-1B
	- lora
	- sft
	- transformers
	- trl
	---

	# Model Card for NPC Brain

	This is a fine-tune of Open BMB's Mini CPM5 1B parameter model.

	I have specifically trained it to be the "Brain" of my app for the Build Small Hackathon in June 2026



	- Developed by: mecoffey
	- Model type: Causal Language Model
	- Language: English
	- License: apache-2.0
	- Finetuned from model: https://huggingface.co/openbmb/MiniCPM5-1B
	- Total Parameters after fine-tune 1,080,632,832



	### Recommendations

	This model was trained with the following workflow in mind:

	- User prompt = physical description of a fantasy character.
	- User prompt + generated description = a timeline backstory for the character to get to know them.

	The user prompt is designed to be short and simple. "something random", "a farmer" "a Half-Orc barkeep with a noticeable scar"


	## Training Details

	### Training Data

	https://huggingface.co/datasets/mecoffey/npc_dataset


	### Compute Infrastructure

	Trained with Modal GPUs

	## Citation

	@article{minicpm4,
	title={Minicpm4: Ultra-efficient llms on end devices},
	author={MiniCPM, Team},
	journal={arXiv preprint arXiv:2506.07900},
	year={2025}
	}