biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

Add full API docs and improve the UX of navigating them (#17485)

Browse Source 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
Harry Mellor
2025-05-04 03:42:43 +01:00
committed by GitHub
parent 46fae69cf0
commit d6484ef3c3
101 changed files with 872 additions and 980 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
vllm/model_executor/models/llava_next.py
View File

@@ -537,7 +537,7 @@ class LlavaNextForConditionalGeneration(nn.Module, SupportsMultiModal,
Unlike in LLaVA-1.5, the number of image tokens inputted to the language
model depends on the original size of the input image. Including the
original image token in the input, the required number of image tokens
is given by :func:`get_llava_next_image_feature_size`.
is given by {func}`get_llava_next_image_feature_size`.
This way, the `positions` and `attn_metadata` are consistent
with the `input_ids`.
@@ -548,8 +548,9 @@ class LlavaNextForConditionalGeneration(nn.Module, SupportsMultiModal,
pixel_values: The pixels in each grid patch for each input image.
image_sizes: The original `(height, width)` for each input image.
See also:
:class:`LlavaNextImageInputs`
:::{seealso}
{class}`LlavaNextImageInputs`
:::
"""
if intermediate_tensors is not None:
inputs_embeds = None