daniel/h3-blackwell-runtime
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Make process_image return 2D grid_thw [[t,h,w]] for multi-image support

Browse source
This commit is contained in:
Daniel Maddern 2026-08-19 22:04:12 +07:00
parent 6866263631
commit 67e4b12246
1 changed files with 1 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
src/h3_blackwell_runtime/qwen3vl_vision.py
View file

@ -103,7 +103,7 @@ def process_image(
grid_h = h_bar // patch_size
grid_w = w_bar // patch_size
grid_thw = torch.tensor([1, grid_h, grid_w], device=device, dtype=torch.long)
grid_thw = torch.tensor([[1, grid_h, grid_w]], device=device, dtype=torch.long)
pixel_values = normalized
channel = pixel_values.shape[0]