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

Add GLM4.1V model (Draft) (#19331)

Browse Source 
Signed-off-by: zRzRzRzRzRzRzR <2448370773@qq.com>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
This commit is contained in:
Yuxuan Zhang
2025-07-01 20:48:26 +08:00
committed by GitHub
parent 650d5dbd04
commit ed70f3c64f
17 changed files with 1946 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

40
examples/offline_inference/vision_language.py
View File

@ -248,6 +248,42 @@ def run_glm4v(questions: list[str], modality: str) -> ModelRequestData:
)
# GLM-4.1V
def run_glm4_1v(questions: list[str], modality: str) -> ModelRequestData:
model_name = "THUDM/GLM-4.1V-9B-Thinking"
engine_args = EngineArgs(
model=model_name,
max_model_len=4096,
max_num_seqs=2,
mm_processor_kwargs={
"size": {"shortest_edge": 12544, "longest_edge": 47040000},
"fps": 1,
},
limit_mm_per_prompt={modality: 1},
enforce_eager=True,
)
if modality == "image":
placeholder = "<|begin_of_image|><|image|><|end_of_image|>"
elif modality == "video":
placeholder = "<|begin_of_video|><|video|><|end_of_video|>"
prompts = [
(
"[gMASK]<sop><|system|>\nYou are a helpful assistant.<|user|>\n"
f"{placeholder}"
f"{question}<|assistant|>assistant\n"
)
for question in questions
]
return ModelRequestData(
engine_args=engine_args,
prompts=prompts,
)
# H2OVL-Mississippi
def run_h2ovl(questions: list[str], modality: str) -> ModelRequestData:
assert modality == "image"
@ -1114,6 +1150,7 @@ model_example_map = {
"fuyu": run_fuyu,
"gemma3": run_gemma3,
"glm4v": run_glm4v,
"glm4_1v": run_glm4_1v,
"h2ovl_chat": run_h2ovl,
"idefics3": run_idefics3,
"internvl_chat": run_internvl,
@ -1172,10 +1209,11 @@ def get_multi_modal_input(args):
if args.modality == "video":
# Input video and question
video = VideoAsset(name="baby_reading", num_frames=args.num_frames).np_ndarrays
metadata = VideoAsset(name="baby_reading", num_frames=args.num_frames).metadata
vid_questions = ["Why is this video funny?"]
return {
"data": video,
"data": [(video, metadata)] if args.model_type == "glm4_1v" else video,
"questions": vid_questions,
}