Yolo Ultralytics stopped exporting fp16 tflite format. Can tflite-server directly handle fp32.tflite models?
-
Hey ModalAI,
Your documentation on custom inference using voxl-tflite-server is quite outdated. Yolo Ultralytics has stopped exporting frozen graph model format to tflite fp16 quantization. Ultralytics says there's no quantize=16 option — an FP32 .tflite model automatically runs in FP16 at runtime when using a GPU delegate (WebGPU/OpenCL/Metal), so a separate FP16 export file isn't needed. That's why half=True/quantize=16 throws the assertion.
Can voxl-tflite-server automatically execute FP32 models internally in FP16?
Hardware: VOXL2
Source code: Latest voxl-tflite-server code on modalai gitlab page.
Reading along? Create a free ModalAI Forum account to join in.
With an account you can reply, ask your own question, get an email when a ModalAI engineer answers, and mark the reply that solved it. Your place in each thread is saved between visits.
Questions about VOXL, Flight Core, ESCs and ModalAI drones are answered here by the engineers who build them.
Register Login