adjust concurrency for 100-160gb

OpenRouterTeam · Jan 3, 2024 · b1ddf4f · b1ddf4f
1 parent 6fab93e
commit b1ddf4f
Showing 1 changed file with 1 addition and 1 deletion.
diff --git a/modal/runner/containers/vllm_unified.py b/modal/runner/containers/vllm_unified.py
@@ -64,5 +64,5 @@ def __init__(
     "VllmContainerA100_80G", num_gpus=1, memory=80
 )
 VllmContainerA100_160G = _make_container(
-    "VllmContainerA100_160G", num_gpus=2, memory=80, concurrent_inputs=4
+    "VllmContainerA100_160G", num_gpus=2, memory=80, concurrent_inputs=2
 )