FastVLM: Efficient Vision Encoding for Vision Language Models
Paper • 2412.13303 • Published • 75
FastVLM
Efficient Vision Encoding for Vision Language Models
-
-
FastVLM WebGPU
🍎445Real-time video captioning powered by FastVLM
-
apple/FastVLM-0.5B
Text Generation • 0.8B • Updated • 14k • 388 -
apple/FastVLM-1.5B
Text Generation • 2B • Updated • 2.65k • 79 -
apple/FastVLM-7B
Text Generation • 8B • Updated • 1.63k • 269
apple/FastVLM-0.5B-fp16
0.6B • Updated • 70 • 23Note MLX checkpoint
apple/FastVLM-1.5B-int8
0.5B • Updated • 93 • 17Note MLX checkpoint
apple/FastVLM-7B-int4
1B • Updated • 150 • 28Note MLX checkpoint