A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 16, 17:00 UTC
[CI/Build] Disable multi-node test for InternVL2 (#8428)
Roger Wang committed
Sep 12, 2024 at 15:05 UTC
b61bd98f907180c70f65e21505b3af6d1cc2bf36
1 file changed
+4
-3
tests/distributed/test_pipeline_parallel.py
+4
-3
@@ -32,9 +32,10 @@ VLLM_MULTI_NODE = os.getenv("VLLM_MULTI_NODE", "0") == "1"
32
(1, 4, 1, 0, 0, "meta-llama/Meta-Llama-3-8B", "ray"),
33
(2, 2, 1, 0, 0, "meta-llama/Meta-Llama-3-8B", "ray"),
34
(2, 2, 0, 1, 0, "meta-llama/Meta-Llama-3-8B", "ray"),
35
- (1, 2, 1, 1, 1, "OpenGVLab/InternVL2-1B", "ray"),
36
- (1, 2, 1, 1, 1, "OpenGVLab/InternVL2-2B", "ray"),
37
- (1, 2, 1, 0, 1, "OpenGVLab/InternVL2-4B", "ray"),
35
+ # TODO: Enable internVL2 in a separate test if needed
36
+ # (1, 2, 1, 1, 1, "OpenGVLab/InternVL2-1B", "ray"),
37
+ # (1, 2, 1, 1, 1, "OpenGVLab/InternVL2-2B", "ray"),
38
+ # (1, 2, 1, 0, 1, "OpenGVLab/InternVL2-4B", "ray"),
39
],
40
)
41
@fork_new_process_for_each_test