EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

Multi-Stage Awake: Support Resume and Pause KV Cache and Weights separately (#7099)

Browse Source
This commit is contained in:
Stefan He
2025-06-19 00:56:37 -07:00
committed by GitHub
parent 9179ea1595
commit 3774f07825
14 changed files with 297 additions and 108 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
test/srt/run_suite.py
View File

@@ -74,7 +74,6 @@ suites = {
TestFile("test_radix_attention.py", 105),
TestFile("test_reasoning_content.py", 89),
TestFile("test_regex_constrained.py", 64),
TestFile("test_release_memory_occupation.py", 44),
TestFile("test_request_length_validation.py", 31),
TestFile("test_retract_decode.py", 54),
TestFile("test_server_args.py", 1),
@@ -146,6 +145,7 @@ suites = {
TestFile("test_patch_torch.py", 19),
TestFile("test_update_weights_from_distributed.py", 103),
TestFile("test_verl_engine_2_gpu.py", 64),
TestFile("test_release_memory_occupation.py", 44),
],
"per-commit-2-gpu-amd": [
TestFile("models/lora/test_lora_tp.py", 116),