apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2026-04-23 08:21:53 +08:00
Code Issues Actions 19 Packages Projects Releases Wiki Activity

[Feature] consider multimodal model when dummy run (#6045)

Browse Source 
* add mm do profile

* updata code

* update code

* update code

* update code

* update test case

* update code

* update code

* fix xpu bug

* update code

* add mm do profile

* update test case

* update code
This commit is contained in:
kevin
2026-02-09 17:49:55 +08:00
committed by GitHub
parent 783d56e28a
commit d60daca4a8
25 changed files with 166 additions and 19 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/graph_optimization/test_cuda_graph_spec_decode.py
+1
View File
@@ -106,6 +106,7 @@ class TestCUDAGrpahSpecDecode(unittest.TestCase):
model_config = Mock()
model_config.max_model_len = 512
model_config.architectures = ["test_model"]
model_config.mm_max_tokens_per_item = None
# Initialize cuda graph capture list
graph_opt_config._set_cudagraph_sizes(max_capture_size=scheduler_config.max_num_seqs)
graph_opt_config.init_with_cudagrpah_size(max_capture_size=scheduler_config.max_num_seqs)