Upload PyTorch INT8 quantized FastVLM-1.5B (weights + config + eval) 610c048 Azaz666 commited on Mar 27