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

[Model] Aya Vision (#15441)

Browse Source 
Signed-off-by: Jennifer Zhao <ai.jenniferzhao@gmail.com>
Signed-off-by: Roger Wang <ywang@roblox.com>
Co-authored-by: Roger Wang <ywang@roblox.com>
This commit is contained in:
Jennifer Zhao
2025-04-01 09:30:43 -07:00
committed by GitHub
parent dfa82e2a3d
commit 38327cf454
10 changed files with 617 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
tests/models/decoder_only/vision_language/test_models.py
View File

@@ -158,6 +158,20 @@ VLM_TEST_SETTINGS = {
max_tokens=64,
marks=[large_gpu_mark(min_gb=64)],
),
"aya_vision": VLMTestInfo(
models=["CohereForAI/aya-vision-8b"],
test_type=(VLMTestType.IMAGE, VLMTestType.MULTI_IMAGE),
prompt_formatter=lambda img_prompt: f"<|START_OF_TURN_TOKEN|><|USER_TOKEN|>{img_prompt}<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>", # noqa: E501
single_image_prompts=IMAGE_ASSETS.prompts({
"stop_sign": "<image>What's the content in the center of the image?", # noqa: E501
"cherry_blossom": "<image>What is the season?", # noqa: E501
}),
multi_image_prompt="<image><image>Describe the two images in detail.", # noqa: E501
max_model_len=8192,
max_num_seqs=2,
auto_cls=AutoModelForImageTextToText,
vllm_runner_kwargs={"mm_processor_kwargs": {"crop_to_patches": True}}
),
"blip2": VLMTestInfo(
# TODO: Change back to 2.7b once head_dim = 80 is supported
models=["Salesforce/blip2-opt-6.7b"],

1
tests/models/multimodal/processing/test_common.py
View File

@@ -246,6 +246,7 @@ def _test_processing_correctness_mistral(
# yapf: disable
@pytest.mark.parametrize("model_id", [
"rhymes-ai/Aria",
"CohereForAI/aya-vision-8b",
"Salesforce/blip2-opt-2.7b",
"facebook/chameleon-7b",
"deepseek-ai/deepseek-vl2-tiny",

1
tests/models/registry.py
View File

@@ -259,6 +259,7 @@ _CROSS_ENCODER_EXAMPLE_MODELS = {
_MULTIMODAL_EXAMPLE_MODELS = {
# [Decoder-only]
"AriaForConditionalGeneration": _HfExamplesInfo("rhymes-ai/Aria"),
"AyaVisionForConditionalGeneration": _HfExamplesInfo("CohereForAI/aya-vision-8b"), # noqa: E501
"Blip2ForConditionalGeneration": _HfExamplesInfo("Salesforce/blip2-opt-2.7b", # noqa: E501
extras={"6b": "Salesforce/blip2-opt-6.7b"}), # noqa: E501
"ChameleonForConditionalGeneration": _HfExamplesInfo("facebook/chameleon-7b"), # noqa: E501