GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
haoquan03
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
haoquan03
Model Tree
Input Modalities
Output Modalities
Supported Functionality
Install a CUDA-compatible PyTorch build, then:
bash
pip install transformers==5.12.1 accelerate pillow torchvisionpython
import torchfrom PIL import Imagefrom transformers import AutoModelForImageTextToText, AutoProcessor model_id = "haoquan03/PixelBench-Qwen3-VL-2B-FV"processor = AutoProcessor.from_pretrained(model_id)model = AutoModelForImageTextToText.from_pretrained( model_id, dtype="auto", device_map="auto", attn_implementation="sdpa").eval() messages = [{"role": "user", "content": [ {"type": "image", "image": Image.open("image.jpg").convert("RGB")}, {"type": "text", "text": "Describe this image."},]}]inputs = processor.apply_chat_template( messages, tokenize=True, add_generation_prompt=True, return_dict=True, return_tensors="pt",).to(model.device) with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16): output = model.generate(**inputs, max_new_tokens=256, do_sample=False, use_cache=True)print(processor.decode(output[0, inputs.input_ids.shape[1]:], skip_special_tokens=True))Training settings and the evaluation protocol are available in the code repository.
bibtex
@article{zhang2026pixelbench, title={Can {LLMs} Understand and Generate Images Pixel by Pixel?}, author={Haoquan Zhang and Feiyang Liu and Juanxi Tian and Wanhua Li and Haodong Duan and Weiyang Liu}, journal={arXiv preprint arXiv:TBA}, year={2026}, url={https://haoquanzhang.github.io/pixelbench/}}import torchfrom PIL import Imagefrom transformers import AutoModelForImageTextToText, AutoProcessor model_id = "haoquan03/PixelBench-Qwen3-VL-2B-FV"processor = AutoProcessor.from_pretrained(model_id)model = AutoModelForImageTextToText.from_pretrained( model_id, dtype="auto", device_map="auto", attn_implementation="sdpa").eval() messages = [{"role": "user", "content": [ {"type": "image", "image": Image.open("image.jpg").convert("RGB")}, {"type": "text", "text": "Describe this image."},]}]inputs = processor.apply_chat_template( messages, tokenize=True, add_generation_prompt=True, return_dict=True, return_tensors="pt",).to(model.device) with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16): output = model.generate(**inputs, max_new_tokens=256, do_sample=False, use_cache=True)print(processor.decode(output[0, inputs.input_ids.shape[1]:], skip_special_tokens=True))