obscura/vllm
2
0
Fork 0
mirror of https://github.com/vllm-project/vllm.git synced 2026-08-23 14:10:14 +00:00
Code Issues Packages Projects Releases Wiki Activity

[CI/Build][AMD] Use float16 in test_reset_prefix_cache_e2e to avoid accuracy issues (#29997)

Browse Source 
Signed-off-by: Randall Smith <[email protected]>
Co-authored-by: Randall Smith <[email protected]>
This commit is contained in:
rasmith
2025-12-05 08:42:25 +00:00
committed by GitHub
co-authored by Randall Smith
parent 6038b1b04b
commit feecba09af
1 changed files with 1 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/v1/core/test_reset_prefix_cache_e2e.py
+1
View File
@@ -21,6 +21,7 @@ def test_reset_prefix_cache_e2e(monkeypatch):
max_num_batched_tokens=32,
max_model_len=2048,
compilation_config={"mode": 0},
dtype="float16",
)
engine = LLMEngine.from_engine_args(engine_args)
sampling_params = SamplingParams(