A high-throughput and memory-efficient inference and serving engine for LLMs
[Bugfix][CI] Fix `test_offloading_connector.py::test_fs_tiering_offloading` (#44903)
Signed-off-by: NickLucche <nlucches@redhat.com>
Nicolò Lucchesi committed
Jun 9, 2026 at 09:57 UTC
dab60fc65894e7dadcc9a126f68b138af839d500
1 file changed
+1
-1
tests/v1/kv_connector/unit/test_offloading_connector.py
+1
-1
@@ -377,7 +377,7 @@ def test_fs_tiering_offloading(tmp_path) -> None:
377
378
llm = LLM(
379
model="meta-llama/Llama-3.2-1B-Instruct",
380
- max_model_len=512,
380
+ max_model_len=4096,
381
gpu_memory_utilization=0.5,
382
kv_events_config=kv_events_config,
383
kv_transfer_config=kv_transfer_config,