From 7b79cc4fe2b247c8697c4e0475c5ac4e775affb4 Mon Sep 17 00:00:00 2001 From: b8zhong Date: Thu, 13 Nov 2025 23:57:33 -0800 Subject: [PATCH] ci: speed up b200 ci (#13237) --- test/srt/test_deepseek_v3_fp4_4gpu.py | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/test/srt/test_deepseek_v3_fp4_4gpu.py b/test/srt/test_deepseek_v3_fp4_4gpu.py index a63f52501..d92216e7a 100644 --- a/test/srt/test_deepseek_v3_fp4_4gpu.py +++ b/test/srt/test_deepseek_v3_fp4_4gpu.py @@ -35,6 +35,8 @@ class TestDeepseekV3FP4(CustomTestCase): "modelopt_fp4", "--kv-cache-dtype", "fp8_e4m3", + "--model-loader-extra-config", + '{"enable_multithread_load": true}', ] cls.process = popen_launch_server( cls.model, @@ -106,6 +108,8 @@ class TestDeepseekV3FP4MTP(CustomTestCase): "4", "--kv-cache-dtype", "fp8_e4m3", + "--model-loader-extra-config", + '{"enable_multithread_load": true}', ] cls.process = popen_launch_server( cls.model, @@ -184,6 +188,8 @@ class TestDeepseekV3FP4CutlassMoE(CustomTestCase): "flashinfer_cutlass", "--quantization", "modelopt_fp4", + "--model-loader-extra-config", + '{"enable_multithread_load": true}', ] cls.process = popen_launch_server( cls.model,