Instructions to use zerofata/MS3.2-PaintedFantasy-v3-24B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use zerofata/MS3.2-PaintedFantasy-v3-24B with Transformers:

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

pipe = pipeline("text-generation", model="zerofata/MS3.2-PaintedFantasy-v3-24B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("zerofata/MS3.2-PaintedFantasy-v3-24B")
model = AutoModelForCausalLM.from_pretrained("zerofata/MS3.2-PaintedFantasy-v3-24B")
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]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use zerofata/MS3.2-PaintedFantasy-v3-24B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "zerofata/MS3.2-PaintedFantasy-v3-24B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "zerofata/MS3.2-PaintedFantasy-v3-24B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/zerofata/MS3.2-PaintedFantasy-v3-24B

SGLang

How to use zerofata/MS3.2-PaintedFantasy-v3-24B 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 "zerofata/MS3.2-PaintedFantasy-v3-24B" \
    --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": "zerofata/MS3.2-PaintedFantasy-v3-24B",
		"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 "zerofata/MS3.2-PaintedFantasy-v3-24B" \
        --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": "zerofata/MS3.2-PaintedFantasy-v3-24B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use zerofata/MS3.2-PaintedFantasy-v3-24B with Docker Model Runner:
```
docker model run hf.co/zerofata/MS3.2-PaintedFantasy-v3-24B
```

Gonna try it!

by Karlmeister - opened Nov 10, 2025

Discussion

Karlmeister

Nov 10, 2025

Being PF 24b v1 & v2 ones of my fav models to rp in the last months, and having tried v3 34b - amazing too but its size limits the available space for context in VRAM, I was looking forward a new 24b version. Can't wait to try it! Thank you very much!

zerofata

Owner Nov 11, 2025

thanks! lemme know what you think.

I like the model, but I'm thinking I need to cut down the verbosity a bit.

Karlmeister

Nov 14, 2025

•

edited Nov 15, 2025

Amazing model. With recomm settings, is highly consistent, fun, creative and yeah, a lil bit verbose but it ain't an issue to me as I use to cut the answer at 768 tokens. Sometimes - prolly by the System prompt context - it's a lil hard to introduce new characters, but after a slight push, it manages them pretty well. And the context size is increased. Now, around 18k tokens it starts to randomly repeat stuff (before was around 12k).

EDIT: Agree with it's a lil bit verbose, sometimes it overwhelms you with several questions. Good thing is that they are consistent and related to the context, but it may be overwhelming (or even annoying for some) 4 questions in the same answer.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment