biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[v1] Hybrid Memory Allocator (#17996)

Browse Source 
Signed-off-by: Chen Zhang <zhangch99@outlook.com>
This commit is contained in:
Chen Zhang
2025-06-06 11:47:09 +08:00
committed by GitHub
parent 3465b87ef8
commit f8a1a2d108
21 changed files with 1605 additions and 440 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
tests/v1/e2e/test_correctness_sliding_window.py
View File

@@ -18,7 +18,7 @@ class TestConfig:
model_config = {
"bigcode/starcoder2-3b": TestConfig(4096, (800, 1100)),
"google/gemma-2-2b-it": TestConfig(4096, (400, 800)),
"google/gemma-3-1b-it": TestConfig(4096, (400, 800)),
}
@@ -26,7 +26,7 @@ model_config = {
"model",
[
"bigcode/starcoder2-3b", # sliding window only
"google/gemma-2-2b-it", # sliding window + full attention
"google/gemma-3-1b-it", # sliding window + full attention
])
@pytest.mark.parametrize("batch_size", [5])
@pytest.mark.parametrize("seed", [1])