[CI] Fix CI timeouts by upgrading runai_model_streamer (related to #16937) (#17636)

.github/workflows/pr-test.yml

@@ -1131,6 +1131,8 @@ jobs:
           CUSTOM_BUILD_SGL_KERNEL=${{needs.check-changes.outputs.sgl_kernel}} bash scripts/ci/cuda/ci_install_dependency.sh diffusion
       - name: Run diffusion server tests
         timeout-minutes: 240
+        env:
+          RUNAI_STREAMER_MEMORY_LIMIT: 0
         run: |
           cd python
           CONTINUE_ON_ERROR_FLAG=""
@@ -1183,6 +1185,8 @@ jobs:
 
       - name: Run diffusion server tests
         timeout-minutes: 240
+        env:
+          RUNAI_STREAMER_MEMORY_LIMIT: 0
         run: |
           cd python
           CONTINUE_ON_ERROR_FLAG=""

python/pyproject.toml

@@ -95,7 +95,7 @@ diffusion = [
   "remote-pdb",
   "st_attn==0.0.7 ; platform_machine != 'aarch64' and platform_machine != 'arm64'",
   "vsa==0.0.4 ; platform_machine != 'aarch64' and platform_machine != 'arm64'",
-  "runai_model_streamer",
+  "runai_model_streamer>=0.15.5",
   "cache-dit==1.2.0"
 ]
 

python/pyproject_other.toml

@@ -96,7 +96,7 @@ diffusion_hip = [
   "torchcodec==0.5.0",
   "st_attn==0.0.7",
   "vsa==0.0.4",
-  "runai_model_streamer",
+  "runai_model_streamer>=0.15.5",
 ]
 
 # For Intel Gaudi(device : hpu) follow the installation guide
@@ -124,7 +124,7 @@ diffusion_musa = [
   "remote-pdb",
   "st_attn==0.0.7",
   "vsa==0.0.4",
-  "runai_model_streamer",
+  "runai_model_streamer>=0.15.5",
   "cache-dit==1.1.8"
 ]