feat: re-work distro-codegen

each *.py file in the various templates now has to use `Provider`s rather than the stringified provider_types in the DistributionTemplate. Adjust that, regenerate all templates, docs, etc.

Signed-off-by: Charlie Doern <cdoern@redhat.com>

llama_stack/templates/meta-reference-gpu/build.yaml

@@ -3,32 +3,50 @@ distribution_spec:
   description: Use Meta Reference for running LLM inference
   providers:
     inference:
-    - inline::meta-reference
+    - provider_id: meta-reference
+      provider_type: inline::meta-reference
     vector_io:
-    - inline::faiss
-    - remote::chromadb
-    - remote::pgvector
+    - provider_id: faiss
+      provider_type: inline::faiss
+    - provider_id: chromadb
+      provider_type: remote::chromadb
+    - provider_id: pgvector
+      provider_type: remote::pgvector
     safety:
-    - inline::llama-guard
+    - provider_id: llama-guard
+      provider_type: inline::llama-guard
     agents:
-    - inline::meta-reference
+    - provider_id: meta-reference
+      provider_type: inline::meta-reference
     telemetry:
-    - inline::meta-reference
+    - provider_id: meta-reference
+      provider_type: inline::meta-reference
     eval:
-    - inline::meta-reference
+    - provider_id: meta-reference
+      provider_type: inline::meta-reference
     datasetio:
-    - remote::huggingface
-    - inline::localfs
+    - provider_id: huggingface
+      provider_type: remote::huggingface
+    - provider_id: localfs
+      provider_type: inline::localfs
     scoring:
-    - inline::basic
-    - inline::llm-as-judge
-    - inline::braintrust
+    - provider_id: basic
+      provider_type: inline::basic
+    - provider_id: llm-as-judge
+      provider_type: inline::llm-as-judge
+    - provider_id: braintrust
+      provider_type: inline::braintrust
     tool_runtime:
-    - remote::brave-search
-    - remote::tavily-search
-    - inline::rag-runtime
-    - remote::model-context-protocol
+    - provider_id: brave-search
+      provider_type: remote::brave-search
+    - provider_id: tavily-search
+      provider_type: remote::tavily-search
+    - provider_id: rag-runtime
+      provider_type: inline::rag-runtime
+    - provider_id: model-context-protocol
+      provider_type: remote::model-context-protocol
 image_type: conda
+image_name: meta-reference-gpu
 additional_pip_packages:
 - aiosqlite
 - sqlalchemy[asyncio]

llama_stack/templates/meta-reference-gpu/meta_reference.py

@@ -25,19 +25,91 @@ from llama_stack.templates.template import DistributionTemplate, RunConfigSettin
 
 def get_distribution_template() -> DistributionTemplate:
     providers = {
-        "inference": ["inline::meta-reference"],
-        "vector_io": ["inline::faiss", "remote::chromadb", "remote::pgvector"],
-        "safety": ["inline::llama-guard"],
-        "agents": ["inline::meta-reference"],
-        "telemetry": ["inline::meta-reference"],
-        "eval": ["inline::meta-reference"],
-        "datasetio": ["remote::huggingface", "inline::localfs"],
-        "scoring": ["inline::basic", "inline::llm-as-judge", "inline::braintrust"],
+        "inference": [
+            Provider(
+                provider_id="meta-reference",
+                provider_type="inline::meta-reference",
+            )
+        ],
+        "vector_io": [
+            Provider(
+                provider_id="faiss",
+                provider_type="inline::faiss",
+            ),
+            Provider(
+                provider_id="chromadb",
+                provider_type="remote::chromadb",
+            ),
+            Provider(
+                provider_id="pgvector",
+                provider_type="remote::pgvector",
+            ),
+        ],
+        "safety": [
+            Provider(
+                provider_id="llama-guard",
+                provider_type="inline::llama-guard",
+            )
+        ],
+        "agents": [
+            Provider(
+                provider_id="meta-reference",
+                provider_type="inline::meta-reference",
+            )
+        ],
+        "telemetry": [
+            Provider(
+                provider_id="meta-reference",
+                provider_type="inline::meta-reference",
+            )
+        ],
+        "eval": [
+            Provider(
+                provider_id="meta-reference",
+                provider_type="inline::meta-reference",
+            )
+        ],
+        "datasetio": [
+            Provider(
+                provider_id="huggingface",
+                provider_type="remote::huggingface",
+            ),
+            Provider(
+                provider_id="localfs",
+                provider_type="inline::localfs",
+            ),
+        ],
+        "scoring": [
+            Provider(
+                provider_id="basic",
+                provider_type="inline::basic",
+            ),
+            Provider(
+                provider_id="llm-as-judge",
+                provider_type="inline::llm-as-judge",
+            ),
+            Provider(
+                provider_id="braintrust",
+                provider_type="inline::braintrust",
+            ),
+        ],
         "tool_runtime": [
-            "remote::brave-search",
-            "remote::tavily-search",
-            "inline::rag-runtime",
-            "remote::model-context-protocol",
+            Provider(
+                provider_id="brave-search",
+                provider_type="remote::brave-search",
+            ),
+            Provider(
+                provider_id="tavily-search",
+                provider_type="remote::tavily-search",
+            ),
+            Provider(
+                provider_id="rag-runtime",
+                provider_type="inline::rag-runtime",
+            ),
+            Provider(
+                provider_id="model-context-protocol",
+                provider_type="remote::model-context-protocol",
+            ),
         ],
     }
     name = "meta-reference-gpu"

llama_stack/templates/meta-reference-gpu/run-with-safety.yaml

@@ -24,7 +24,6 @@ providers:
       max_seq_len: ${env.MAX_SEQ_LEN:=4096}
   - provider_id: sentence-transformers
     provider_type: inline::sentence-transformers
-    config: {}
   - provider_id: meta-reference-safety
     provider_type: inline::meta-reference
     config:
@@ -88,10 +87,8 @@ providers:
   scoring:
   - provider_id: basic
     provider_type: inline::basic
-    config: {}
   - provider_id: llm-as-judge
     provider_type: inline::llm-as-judge
-    config: {}
   - provider_id: braintrust
     provider_type: inline::braintrust
     config:
@@ -109,10 +106,8 @@ providers:
       max_results: 3
   - provider_id: rag-runtime
     provider_type: inline::rag-runtime
-    config: {}
   - provider_id: model-context-protocol
     provider_type: remote::model-context-protocol
-    config: {}
 metadata_store:
   type: sqlite
   db_path: ${env.SQLITE_STORE_DIR:=~/.llama/distributions/meta-reference-gpu}/registry.db

llama_stack/templates/meta-reference-gpu/run.yaml

@@ -24,7 +24,6 @@ providers:
       max_seq_len: ${env.MAX_SEQ_LEN:=4096}
   - provider_id: sentence-transformers
     provider_type: inline::sentence-transformers
-    config: {}
   vector_io:
   - provider_id: faiss
     provider_type: inline::faiss
@@ -78,10 +77,8 @@ providers:
   scoring:
   - provider_id: basic
     provider_type: inline::basic
-    config: {}
   - provider_id: llm-as-judge
     provider_type: inline::llm-as-judge
-    config: {}
   - provider_id: braintrust
     provider_type: inline::braintrust
     config:
@@ -99,10 +96,8 @@ providers:
       max_results: 3
   - provider_id: rag-runtime
     provider_type: inline::rag-runtime
-    config: {}
   - provider_id: model-context-protocol
     provider_type: remote::model-context-protocol
-    config: {}
 metadata_store:
   type: sqlite
   db_path: ${env.SQLITE_STORE_DIR:=~/.llama/distributions/meta-reference-gpu}/registry.db