Unverified Commit 9bc6d98d authored by Yuewei Na's avatar Yuewei Na Committed by GitHub
Browse files

fix(test): unskip test_smoke by switching to trtllm attention backend (DYN-2607) (#8104)


Signed-off-by: default avatarYuewei Na <nv-yna@users.noreply.github.com>
Co-authored-by: default avatarYuewei Na <nv-yna@users.noreply.github.com>
parent 1ebe7060
...@@ -6,4 +6,5 @@ kv_cache_config: ...@@ -6,4 +6,5 @@ kv_cache_config:
enable_partial_reuse: false enable_partial_reuse: false
free_gpu_memory_fraction: 0.80 free_gpu_memory_fraction: 0.80
max_tokens: 8192 max_tokens: 8192
compile_backend: torch-cudagraph attn_backend: trtllm
max_batch_size: 128
...@@ -166,7 +166,6 @@ def send_completion_request( ...@@ -166,7 +166,6 @@ def send_completion_request(
raise raise
@pytest.mark.skip(reason="Nightly CI failure: https://linear.app/nvidia/issue/DYN-2607")
@pytest.mark.trtllm @pytest.mark.trtllm
@pytest.mark.e2e @pytest.mark.e2e
@pytest.mark.slow @pytest.mark.slow
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment