LCORE-633: bump-up Llama Stack version to 0.2.19 (#520)

* LCORE-633: bump-up Llama Stack version to 0.2.19

* Update range of supported Llama Stack versions

* Updated run.yaml file

* Fixed e2e test settings

Author: tisnik

.github/workflows/e2e_tests.yaml

@@ -171,7 +171,7 @@ jobs:
                   distributed_backend: null
                   dpo_output_dir: '.'
                 provider_id: huggingface
-                provider_type: inline::huggingface
+                provider_type: inline::huggingface-gpu
               safety:
               - config:
                   excluded_categories: []

pyproject.toml

@@ -28,8 +28,8 @@ dependencies = [
     # Used by auth/k8s integration
     "kubernetes>=30.1.0",
     # Used to call Llama Stack APIs
-    "llama-stack==0.2.18",
-    "llama-stack-client==0.2.18",
+    "llama-stack==0.2.19",
+    "llama-stack-client==0.2.19",
     # Used by Logger
     "rich>=14.0.0",
     # Used by JWK token auth handler

run.yaml

@@ -71,7 +71,7 @@ providers:
       distributed_backend: null
       dpo_output_dir: "."
     provider_id: huggingface
-    provider_type: inline::huggingface
+    provider_type: inline::huggingface-gpu
   safety:
   - config:
       excluded_categories: []

src/constants.py

@@ -2,7 +2,7 @@
 
 # Minimal and maximal supported Llama Stack version
 MINIMAL_SUPPORTED_LLAMA_STACK_VERSION = "0.2.17"
-MAXIMAL_SUPPORTED_LLAMA_STACK_VERSION = "0.2.18"
+MAXIMAL_SUPPORTED_LLAMA_STACK_VERSION = "0.2.19"
 
 UNABLE_TO_PROCESS_RESPONSE = "Unable to process this request"