Instructions to use RaspizdAI/femboyAV-4b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use RaspizdAI/femboyAV-4b with Transformers:

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

pipe = pipeline("text-generation", model="RaspizdAI/femboyAV-4b")
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("RaspizdAI/femboyAV-4b")
model = AutoModelForMultimodalLM.from_pretrained("RaspizdAI/femboyAV-4b", 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 RaspizdAI/femboyAV-4b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "RaspizdAI/femboyAV-4b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "RaspizdAI/femboyAV-4b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/RaspizdAI/femboyAV-4b

SGLang

How to use RaspizdAI/femboyAV-4b 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 "RaspizdAI/femboyAV-4b" \
    --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": "RaspizdAI/femboyAV-4b",
		"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 "RaspizdAI/femboyAV-4b" \
        --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": "RaspizdAI/femboyAV-4b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use RaspizdAI/femboyAV-4b with Docker Model Runner:
```
docker model run hf.co/RaspizdAI/femboyAV-4b
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

femboyAV-4b

A LoRA fine-tune of huihui-ai/Huihui-Qwen3-VL-4B-Instruct-abliterated on 10,000 English/Russian chat pairs with a playful assistant persona (emoticons like :3, UwU, OwO, ^^, ~).

Base: Qwen3-VL-4B-Instruct (abliterated), ~4.48B parameters
Data: RaspizdAI/femboy-10k (10,000 public user_input / ai_output pairs; training used the full 10,330-pair set), 1 epoch
Training: Unsloth LoRA, 39,321,600 trainable parameters of 4,477,137,408 (0.88%), lr 2e-4 peak
Vision: the vision tower was not modified, so image inputs still work

GGUF builds (F16, Q8_0, Q4_K_M and the vision projector) are available in RaspizdAI/femboyAV-4b-gguf.

Files

file	description
`model.safetensors`	merged LoRA weights (BF16), ~8.9 GB
`config.json`	model config
`generation_config.json`	generation defaults
`tokenizer.json`, `tokenizer_config.json`	tokenizer
`chat_template.jinja`	chat template
`preprocessor_config.json`, `video_preprocessor_config.json`	image/video preprocessing

Usage

from transformers import AutoProcessor, AutoModelForImageTextToText
import torch

model_id = "RaspizdAI/femboyAV-4b"
processor = AutoProcessor.from_pretrained(model_id)
model = AutoModelForImageTextToText.from_pretrained(model_id, torch_dtype=torch.bfloat16, device_map="auto")

messages = [{"role": "user", "content": [{"type": "text", "text": "Привет! Как дела?"}]}]
inputs = processor.apply_chat_template(messages, add_generation_prompt=True, tokenize=True,
                                       return_dict=True, return_tensors="pt").to(model.device)
out = model.generate(**inputs, max_new_tokens=128, do_sample=True, temperature=0.8)
print(processor.batch_decode(out[:, inputs["input_ids"].shape[1]:], skip_special_tokens=True)[0])

Limitations

The persona is stylistic; replies are not factual advice and may contain invented details.
Trained on synthetic data; may produce repetitive or overly emotive text.
Content is informal; not intended for professional, medical, legal or safety-critical use.

License

Apache-2.0, following the base model.

Downloads last month: -

Safetensors

Model size

4B params

Tensor type

BF16

Model tree for RaspizdAI/femboyAV-4b

Base model

Qwen/Qwen3-VL-4B-Instruct

Finetuned

huihui-ai/Huihui-Qwen3-VL-4B-Instruct-abliterated

Finetuned

(8)

this model

Quantizations

1 model

RaspizdAI
/

femboyAV-4b