add export to FP16

add inference with possibility to have different
2026-04-23 01:06:35 +00:00 · 2025-03-28 12:54:25 +02:00
parent eaef1a9b66
commit 5b89a21b36
9 changed files with 365 additions and 242 deletions
@@ -16,4 +16,5 @@ boto3
 msgpack
 rstream
 onnxruntime-gpu
-netron
+netron
+pycuda