[Model] Enable native HF format InternVL support (#23742)

Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
This commit is contained in:
Isotr0py
2025-08-27 22:45:17 +08:00
committed by GitHub
parent 3af47c3cc6
commit 841490434a
4 changed files with 18 additions and 16 deletions

View File

@@ -222,21 +222,6 @@ VLM_TEST_SETTINGS = {
},
marks=[large_gpu_mark(min_gb=32)],
),
# Check "auto" with fallback to transformers
"internvl-transformers": VLMTestInfo(
models=["OpenGVLab/InternVL3-1B-hf"],
test_type=(VLMTestType.IMAGE, VLMTestType.MULTI_IMAGE),
prompt_formatter=lambda img_prompt: f"<|im_start|>User\n{img_prompt}<|im_end|>\n<|im_start|>Assistant\n", # noqa: E501
img_idx_to_prompt=lambda idx: "<IMG_CONTEXT>",
max_model_len=4096,
use_tokenizer_eos=True,
image_size_factors=[(0.25, 0.5, 1.0)],
vllm_runner_kwargs={
"model_impl": "auto",
},
auto_cls=AutoModelForImageTextToText,
marks=[pytest.mark.core_model],
),
#### Extended model tests
"aria": VLMTestInfo(
models=["rhymes-ai/Aria"],
@@ -461,6 +446,20 @@ VLM_TEST_SETTINGS = {
use_tokenizer_eos=True,
patch_hf_runner=model_utils.internvl_patch_hf_runner,
),
"intern_vl-hf": VLMTestInfo(
models=["OpenGVLab/InternVL3-1B-hf"],
test_type=(
VLMTestType.IMAGE,
VLMTestType.MULTI_IMAGE,
VLMTestType.VIDEO,
),
prompt_formatter=lambda img_prompt: f"<|im_start|>User\n{img_prompt}<|im_end|>\n<|im_start|>Assistant\n", # noqa: E501
img_idx_to_prompt=lambda idx: "<IMG_CONTEXT>",
video_idx_to_prompt=lambda idx: "<video>",
max_model_len=8192,
use_tokenizer_eos=True,
auto_cls=AutoModelForImageTextToText,
),
"kimi_vl": VLMTestInfo(
models=["moonshotai/Kimi-VL-A3B-Instruct"],
test_type=(VLMTestType.IMAGE, VLMTestType.MULTI_IMAGE),