import torch
import transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
from PIL import Image
import warnings

# disable some warnings
transformers.logging.set_verbosity_error()
transformers.logging.disable_progress_bar()
warnings.filterwarnings('ignore')

# set device
torch.set_default_device('cuda')  # or 'cpu'

# create model
model = AutoModelForCausalLM.from_pretrained(
    'qnguyen3/nanoLLaVA',
    torch_dtype=torch.float16,
    device_map='auto',
    trust_remote_code=True)
tokenizer = AutoTokenizer.from_pretrained(
    'qnguyen3/nanoLLaVA',
    trust_remote_code=True)

# text prompt
prompt = 'Describe this image in detail'

messages = [
    {"role": "user", "content": f'<image>\n{prompt}'}
]
text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True
)

print(text)

text_chunks = [tokenizer(chunk).input_ids for chunk in text.split('<image>')]
input_ids = torch.tensor(text_chunks[0] + [-200] + text_chunks[1], dtype=torch.long).unsqueeze(0)

# image, sample images can be found in images folder
image = Image.open('/path/to/image.png')
image_tensor = model.process_images([image], model.config).to(dtype=model.dtype)

# generate
output_ids = model.generate(
    input_ids,
    images=image_tensor,
    max_new_tokens=2048,
    use_cache=True)[0]

print(tokenizer.decode(output_ids[input_ids.shape[1]:], skip_special_tokens=True).strip())

Prompt Format

Base LLM: <a href="https://huggingface.co/vilm/Quyen-SE-v0.1" target="_blank" rel="noopener noreferrer" class="text-blue-400 hover:text-blue-300 underline underline-offset-2 transition-colors">Quyen-SE-v0.1 (Qwen1.5-0.5B)
Vision Encoder: <a href="https://huggingface.co/google/siglip-so400m-patch14-384" target="_blank" rel="noopener noreferrer" class="text-blue-400 hover:text-blue-300 underline underline-offset-2 transition-colors">google/siglip-so400m-patch14-384

The model follow the ChatML standard, however, without \n at the end of <|im_end|>:

<|im_start|>system
Answer the question<|im_end|><|im_start|>user
<image>
What is the picture about?<|im_end|><|im_start|>assistant

Image	Example
small	What is the text saying? "Small but mighty". How does the text correlate to the context of the image? The text seems to be a playful or humorous representation of a small but mighty figure, possibly a mouse or a mouse toy, holding a weightlifting bar.

Model is trained using a modified version from Bunny