Instructions to use oldpepper12/PauloGPT-9B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use oldpepper12/PauloGPT-9B with Transformers:

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

pipe = pipeline("image-text-to-text", model="oldpepper12/PauloGPT-9B")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("oldpepper12/PauloGPT-9B")
model = AutoModelForMultimodalLM.from_pretrained("oldpepper12/PauloGPT-9B", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.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=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use oldpepper12/PauloGPT-9B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "oldpepper12/PauloGPT-9B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "oldpepper12/PauloGPT-9B",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/oldpepper12/PauloGPT-9B

SGLang

How to use oldpepper12/PauloGPT-9B 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 "oldpepper12/PauloGPT-9B" \
    --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": "oldpepper12/PauloGPT-9B",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "oldpepper12/PauloGPT-9B" \
        --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": "oldpepper12/PauloGPT-9B",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use oldpepper12/PauloGPT-9B with Docker Model Runner:
```
docker model run hf.co/oldpepper12/PauloGPT-9B
```

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.

PauloGPT-9B

PauloGPT-9B is the first release in the PauloGPT series of frontier language models, and our most capable model to date. It is a 9B (billion) parameter model trained on 4.7M tokens from Paulo 99.242.66.177 2456's server, zero tokens from FineWeb-Edu, and unlike other open-weight models, it is not distilled from Claude.

Key Features

Adjustable intelligence: PauloGPT, being modeled on the twenty-two members of Paulo's server, therefore has 22 effort modes ranging from paulo to mango.10000, controlling both the personality as well as the accuracy of the model's outputs accordingly.
Coding: PauloGPT is a capable coding model. Being trained on real-world computer science students, it excels on narrowly-scoped, fairly esoteric, and highly verifiable programming tasks which leverage textbook data structures and algorithms techniques.
Alignment: PauloGPT is our most aligned model to date. On our internal benchmark, we see that PauloGPT refuses harmful queries 99.93% of the time, such as those iliciting praise of the University of Waterloo.

PauloBench

SoTA score on our internal benchmark, PauloBench, PauloGPT scores over 50%, scaling with effort.

Intelligence Level	Score
paulo	109/140 (78%)
oldpepper12	95/140 (68%)
notnishi.	77/140 (55%)
mango.10000	70/140 (50%)

PauloBench consists of the following fourteen curated questions from domain-experts:

#	Category	Question
1	`arithmetic`	whats 17 * 23
2	`arithmetic`	smallest prime bigger than 100?
3	`word problem`	train leaves at 3:40pm and the trip is 2h 35m, when does it get in
4	`word problem`	bat and ball cost $1.10 total, bat costs $1 more than the ball. how much is the ball
5	`linguistics`	how many r's are in strawberry
6	`calculus`	whats the integral of 1/x
7	`calculus`	derivative of e^(x^2)?
8	`cs theory`	whats the worst case time complexity of heapsort
9	`chemistry`	chemical formula for sulfuric acid?
10	`physics`	roughly whats the speed of light in m/s
11	`history`	who was the first roman emperor
12	`history`	what year did the berlin wall fall
13	`geography`	whats the capital of kazakhstan
14	`math history`	who proved fermat's last theorem

Safety

Due to its advanced capabilities, PauloGPT has been deployed with robust safeguards on deployment contexts such as Agentic League of Legends and queries related to Culinary Arts such as formulating recipes or advising on macronutrients.

Supported personas

Persona	Messages
notnishi.	61,614
redfloorboard	31,737
paulo99.242.66.1772456	28,034
vai9er	27,993
oldpepper12	17,582
mango.10000	15,641
._phone	14,202
vango0	9,948
avltg	8,697
whombst	6,627
aloe.ai	6,114
meeddle	6,078
defaultedplayer	5,098
wooper.	5,094
bigthicctoucan	4,758
caspian9	2,262
mredisonthebest	2,233
lithid	1,847
dead_meme2839	1,089

Chat template

<|im_start|>system
Discord group chat<|im_end|>
<|im_start|>vai9er
whats 17 * 23<|im_end|>
<|im_start|>mango.10000
> whats 17 * 23
391<|im_end|>
<|im_start|>oldpepper12

Downloads last month: 19

Safetensors

Model size

9B params

Tensor type

BF16