Skip to content

[Misc] Avoid use of deprecated AutoModelForVision2Seq #25065

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Sep 17, 2025
+7 −7
Merged

[Misc] Avoid use of deprecated AutoModelForVision2Seq #25065

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 7 additions & 7 deletions tests/models/multimodal/generation/test_common.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,7 +10,7 @@

import pytest
from transformers import (AutoModel, AutoModelForImageTextToText,
AutoModelForTextToWaveform, AutoModelForVision2Seq)
AutoModelForTextToWaveform)

from vllm.platforms import current_platform
from vllm.utils import identity
Expand Down Expand Up @@ -137,7 +137,7 @@
video_idx_to_prompt=lambda idx: "<|vision_start|><|video_pad|><|vision_end|>", # noqa: E501
max_model_len=4096,
max_num_seqs=2,
auto_cls=AutoModelForVision2Seq,
auto_cls=AutoModelForImageTextToText,
vllm_output_post_proc=model_utils.qwen2_vllm_to_hf_output,
image_size_factors=[(), (0.25,), (0.25, 0.25, 0.25), (0.25, 0.2, 0.15)],
marks=[pytest.mark.core_model, pytest.mark.cpu_model],
Expand Down Expand Up @@ -502,7 +502,7 @@
num_video_frames=16,
max_model_len=16384,
hf_model_kwargs=model_utils.llava_onevision_hf_model_kwargs("llava-hf/llava-onevision-qwen2-0.5b-ov-hf"), # noqa: E501
auto_cls=AutoModelForVision2Seq,
auto_cls=AutoModelForImageTextToText,
vllm_output_post_proc=model_utils.llava_onevision_vllm_to_hf_output,
custom_test_opts=[CustomTestOptions(
inputs=custom_inputs.multi_video_multi_aspect_ratio_inputs(
Expand All @@ -518,7 +518,7 @@
num_video_frames=16,
max_model_len=4096,
max_num_seqs=2,
auto_cls=AutoModelForVision2Seq,
auto_cls=AutoModelForImageTextToText,
vllm_output_post_proc=model_utils.llava_video_vllm_to_hf_output,
),
"mantis": VLMTestInfo(
Expand Down Expand Up @@ -680,7 +680,7 @@
multi_image_prompt="Picture 1: <vlm_image>\nPicture 2: <vlm_image>\nDescribe these two images with one paragraph respectively.", # noqa: E501
max_model_len=4096,
max_num_seqs=2,
auto_cls=AutoModelForVision2Seq,
auto_cls=AutoModelForImageTextToText,
vllm_output_post_proc=model_utils.qwen2_vllm_to_hf_output,
image_size_factors=[(), (0.25,), (0.25, 0.25, 0.25), (0.25, 0.2, 0.15)],
marks=[pytest.mark.cpu_model],
Expand Down Expand Up @@ -784,7 +784,7 @@
test_type=VLMTestType.CUSTOM_INPUTS,
max_model_len=16384,
max_num_seqs=2,
auto_cls=AutoModelForVision2Seq,
auto_cls=AutoModelForImageTextToText,
hf_model_kwargs=model_utils.llava_onevision_hf_model_kwargs("llava-hf/llava-onevision-qwen2-0.5b-ov-hf"), # noqa: E501
vllm_output_post_proc=model_utils.llava_onevision_vllm_to_hf_output,
custom_test_opts=[CustomTestOptions(
Expand All @@ -800,7 +800,7 @@
test_type=VLMTestType.CUSTOM_INPUTS,
max_model_len=4096,
max_num_seqs=2,
auto_cls=AutoModelForVision2Seq,
auto_cls=AutoModelForImageTextToText,
vllm_output_post_proc=model_utils.qwen2_vllm_to_hf_output,
custom_test_opts=[CustomTestOptions(
inputs=custom_inputs.windows_attention_image_qwen2_5_vl(),
Expand Down
Loading