Introduce Stable LoRA ID System for Overlapped Updates and Prefix Caching (#8261)

This commit is contained in:
Lifu Huang
2025-07-23 00:32:16 -07:00
committed by GitHub
parent e885bfdc6a
commit 8abd3e77fe
11 changed files with 400 additions and 261 deletions

View File

@@ -14,7 +14,7 @@ class TestFile:
suites = {
"per-commit": [
TestFile("models/lora/test_lora.py", 200),
TestFile("models/lora/test_lora_eviction.py", 120),
TestFile("models/lora/test_lora_eviction.py", 200),
TestFile("models/lora/test_lora_backend.py", 99),
TestFile("models/lora/test_multi_lora_backend.py", 60),
TestFile("models/lora/test_lora_cuda_graph.py", 250),