fix: register provider model name and HF alias in run.yaml (#1304)

Each model known to the system has two identifiers: 

- the `provider_resource_id` (what the provider calls it) -- e.g.,
`accounts/fireworks/models/llama-v3p1-8b-instruct`
- the `identifier` (`model_id`) under which it is registered and gets
routed to the appropriate provider.

We have so far used the HuggingFace repo alias as the standardized
identifier you can use to refer to the model. So in the above example,
we'd use `meta-llama/Llama-3.1-8B-Instruct` as the name under which it
gets registered. This makes it convenient for users to refer to these
models across providers.

However, we forgot to register the _actual_ provider model ID also. You
should be able to route via `provider_resource_id` also, of course.

This change fixes this (somewhat grave) omission.

*Note*: this change is additive -- more aliases work now compared to
before.

## Test Plan

Run the following for distro=(ollama fireworks together)
```
LLAMA_STACK_CONFIG=$distro \
   pytest -s -v tests/client-sdk/inference/test_text_inference.py \
   --inference-model=meta-llama/Llama-3.1-8B-Instruct --vision-inference-model=""
```

docs/source/distributions/remote_hosted_distro/nvidia.md

@@ -27,19 +27,19 @@ The following environment variables can be configured:
 
 The following models are available by default:
 
-- `meta-llama/Llama-3-8B-Instruct (meta/llama3-8b-instruct)`
+- `meta/llama3-8b-instruct (aliases: meta-llama/Llama-3-8B-Instruct)`
-- `meta-llama/Llama-3-70B-Instruct (meta/llama3-70b-instruct)`
+- `meta/llama3-70b-instruct (aliases: meta-llama/Llama-3-70B-Instruct)`
-- `meta-llama/Llama-3.1-8B-Instruct (meta/llama-3.1-8b-instruct)`
+- `meta/llama-3.1-8b-instruct (aliases: meta-llama/Llama-3.1-8B-Instruct)`
-- `meta-llama/Llama-3.1-70B-Instruct (meta/llama-3.1-70b-instruct)`
+- `meta/llama-3.1-70b-instruct (aliases: meta-llama/Llama-3.1-70B-Instruct)`
-- `meta-llama/Llama-3.1-405B-Instruct-FP8 (meta/llama-3.1-405b-instruct)`
+- `meta/llama-3.1-405b-instruct (aliases: meta-llama/Llama-3.1-405B-Instruct-FP8)`
-- `meta-llama/Llama-3.2-1B-Instruct (meta/llama-3.2-1b-instruct)`
+- `meta/llama-3.2-1b-instruct (aliases: meta-llama/Llama-3.2-1B-Instruct)`
-- `meta-llama/Llama-3.2-3B-Instruct (meta/llama-3.2-3b-instruct)`
+- `meta/llama-3.2-3b-instruct (aliases: meta-llama/Llama-3.2-3B-Instruct)`
-- `meta-llama/Llama-3.2-11B-Vision-Instruct (meta/llama-3.2-11b-vision-instruct)`
+- `meta/llama-3.2-11b-vision-instruct (aliases: meta-llama/Llama-3.2-11B-Vision-Instruct)`
-- `meta-llama/Llama-3.2-90B-Vision-Instruct (meta/llama-3.2-90b-vision-instruct)`
+- `meta/llama-3.2-90b-vision-instruct (aliases: meta-llama/Llama-3.2-90B-Vision-Instruct)`
-- `nvidia/llama-3.2-nv-embedqa-1b-v2 (nvidia/llama-3.2-nv-embedqa-1b-v2)`
+- `nvidia/llama-3.2-nv-embedqa-1b-v2 `
-- `nvidia/nv-embedqa-e5-v5 (nvidia/nv-embedqa-e5-v5)`
+- `nvidia/nv-embedqa-e5-v5 `
-- `nvidia/nv-embedqa-mistral-7b-v2 (nvidia/nv-embedqa-mistral-7b-v2)`
+- `nvidia/nv-embedqa-mistral-7b-v2 `
-- `snowflake/arctic-embed-l (snowflake/arctic-embed-l)`
+- `snowflake/arctic-embed-l `
 
 
 ### Prerequisite: API Keys

docs/source/distributions/self_hosted_distro/bedrock.md

@@ -34,9 +34,9 @@ The following environment variables can be configured:
 
 The following models are available by default:
 
-- `meta-llama/Llama-3.1-8B-Instruct (meta.llama3-1-8b-instruct-v1:0)`
+- `meta.llama3-1-8b-instruct-v1:0 (aliases: meta-llama/Llama-3.1-8B-Instruct)`
-- `meta-llama/Llama-3.1-70B-Instruct (meta.llama3-1-70b-instruct-v1:0)`
+- `meta.llama3-1-70b-instruct-v1:0 (aliases: meta-llama/Llama-3.1-70B-Instruct)`
-- `meta-llama/Llama-3.1-405B-Instruct-FP8 (meta.llama3-1-405b-instruct-v1:0)`
+- `meta.llama3-1-405b-instruct-v1:0 (aliases: meta-llama/Llama-3.1-405B-Instruct-FP8)`
 
 
 ### Prerequisite: API Keys

docs/source/distributions/self_hosted_distro/cerebras.md

@@ -27,8 +27,8 @@ The following environment variables can be configured:
 
 The following models are available by default:
 
-- `meta-llama/Llama-3.1-8B-Instruct (llama3.1-8b)`
+- `llama3.1-8b (aliases: meta-llama/Llama-3.1-8B-Instruct)`
-- `meta-llama/Llama-3.3-70B-Instruct (llama-3.3-70b)`
+- `llama-3.3-70b (aliases: meta-llama/Llama-3.3-70B-Instruct)`
 
 
 ### Prerequisite: API Keys

docs/source/distributions/self_hosted_distro/fireworks.md

@@ -37,17 +37,17 @@ The following environment variables can be configured:
 
 The following models are available by default:
 
-- `meta-llama/Llama-3.1-8B-Instruct (accounts/fireworks/models/llama-v3p1-8b-instruct)`
+- `accounts/fireworks/models/llama-v3p1-8b-instruct (aliases: meta-llama/Llama-3.1-8B-Instruct)`
-- `meta-llama/Llama-3.1-70B-Instruct (accounts/fireworks/models/llama-v3p1-70b-instruct)`
+- `accounts/fireworks/models/llama-v3p1-70b-instruct (aliases: meta-llama/Llama-3.1-70B-Instruct)`
-- `meta-llama/Llama-3.1-405B-Instruct-FP8 (accounts/fireworks/models/llama-v3p1-405b-instruct)`
+- `accounts/fireworks/models/llama-v3p1-405b-instruct (aliases: meta-llama/Llama-3.1-405B-Instruct-FP8)`
-- `meta-llama/Llama-3.2-1B-Instruct (accounts/fireworks/models/llama-v3p2-1b-instruct)`
+- `accounts/fireworks/models/llama-v3p2-1b-instruct (aliases: meta-llama/Llama-3.2-1B-Instruct)`
-- `meta-llama/Llama-3.2-3B-Instruct (accounts/fireworks/models/llama-v3p2-3b-instruct)`
+- `accounts/fireworks/models/llama-v3p2-3b-instruct (aliases: meta-llama/Llama-3.2-3B-Instruct)`
-- `meta-llama/Llama-3.2-11B-Vision-Instruct (accounts/fireworks/models/llama-v3p2-11b-vision-instruct)`
+- `accounts/fireworks/models/llama-v3p2-11b-vision-instruct (aliases: meta-llama/Llama-3.2-11B-Vision-Instruct)`
-- `meta-llama/Llama-3.2-90B-Vision-Instruct (accounts/fireworks/models/llama-v3p2-90b-vision-instruct)`
+- `accounts/fireworks/models/llama-v3p2-90b-vision-instruct (aliases: meta-llama/Llama-3.2-90B-Vision-Instruct)`
-- `meta-llama/Llama-3.3-70B-Instruct (accounts/fireworks/models/llama-v3p3-70b-instruct)`
+- `accounts/fireworks/models/llama-v3p3-70b-instruct (aliases: meta-llama/Llama-3.3-70B-Instruct)`
-- `meta-llama/Llama-Guard-3-8B (accounts/fireworks/models/llama-guard-3-8b)`
+- `accounts/fireworks/models/llama-guard-3-8b (aliases: meta-llama/Llama-Guard-3-8B)`
-- `meta-llama/Llama-Guard-3-11B-Vision (accounts/fireworks/models/llama-guard-3-11b-vision)`
+- `accounts/fireworks/models/llama-guard-3-11b-vision (aliases: meta-llama/Llama-Guard-3-11B-Vision)`
-- `nomic-ai/nomic-embed-text-v1.5 (nomic-ai/nomic-embed-text-v1.5)`
+- `nomic-ai/nomic-embed-text-v1.5 `
 
 
 ### Prerequisite: API Keys

docs/source/distributions/self_hosted_distro/groq.md

@@ -37,11 +37,11 @@ The following environment variables can be configured:
 
 The following models are available by default:
 
-- `meta-llama/Llama-3.1-8B-Instruct (groq/llama3-8b-8192)`
+- `groq/llama3-8b-8192 (aliases: meta-llama/Llama-3.1-8B-Instruct)`
-- `meta-llama/Llama-3.1-8B-Instruct (groq/llama-3.1-8b-instant)`
+- `groq/llama-3.1-8b-instant `
-- `meta-llama/Llama-3-70B-Instruct (groq/llama3-70b-8192)`
+- `groq/llama3-70b-8192 (aliases: meta-llama/Llama-3-70B-Instruct)`
-- `meta-llama/Llama-3.3-70B-Instruct (groq/llama-3.3-70b-versatile)`
+- `groq/llama-3.3-70b-versatile (aliases: meta-llama/Llama-3.3-70B-Instruct)`
-- `meta-llama/Llama-3.2-3B-Instruct (groq/llama-3.2-3b-preview)`
+- `groq/llama-3.2-3b-preview (aliases: meta-llama/Llama-3.2-3B-Instruct)`
 
 
 ### Prerequisite: API Keys

docs/source/distributions/self_hosted_distro/sambanova.md

@@ -34,15 +34,15 @@ The following environment variables can be configured:
 
 The following models are available by default:
 
-- `meta-llama/Llama-3.1-8B-Instruct (Meta-Llama-3.1-8B-Instruct)`
+- `Meta-Llama-3.1-8B-Instruct (aliases: meta-llama/Llama-3.1-8B-Instruct)`
-- `meta-llama/Llama-3.1-70B-Instruct (Meta-Llama-3.1-70B-Instruct)`
+- `Meta-Llama-3.1-70B-Instruct (aliases: meta-llama/Llama-3.1-70B-Instruct)`
-- `meta-llama/Llama-3.1-405B-Instruct-FP8 (Meta-Llama-3.1-405B-Instruct)`
+- `Meta-Llama-3.1-405B-Instruct (aliases: meta-llama/Llama-3.1-405B-Instruct-FP8)`
-- `meta-llama/Llama-3.2-1B-Instruct (Meta-Llama-3.2-1B-Instruct)`
+- `Meta-Llama-3.2-1B-Instruct (aliases: meta-llama/Llama-3.2-1B-Instruct)`
-- `meta-llama/Llama-3.2-3B-Instruct (Meta-Llama-3.2-3B-Instruct)`
+- `Meta-Llama-3.2-3B-Instruct (aliases: meta-llama/Llama-3.2-3B-Instruct)`
-- `meta-llama/Llama-3.3-70B-Instruct (Meta-Llama-3.3-70B-Instruct)`
+- `Meta-Llama-3.3-70B-Instruct (aliases: meta-llama/Llama-3.3-70B-Instruct)`
-- `meta-llama/Llama-3.2-11B-Vision-Instruct (Llama-3.2-11B-Vision-Instruct)`
+- `Llama-3.2-11B-Vision-Instruct (aliases: meta-llama/Llama-3.2-11B-Vision-Instruct)`
-- `meta-llama/Llama-3.2-90B-Vision-Instruct (Llama-3.2-90B-Vision-Instruct)`
+- `Llama-3.2-90B-Vision-Instruct (aliases: meta-llama/Llama-3.2-90B-Vision-Instruct)`
-- `meta-llama/Llama-Guard-3-8B (Meta-Llama-Guard-3-8B)`
+- `Meta-Llama-Guard-3-8B (aliases: meta-llama/Llama-Guard-3-8B)`
 
 
 ### Prerequisite: API Keys

docs/source/distributions/self_hosted_distro/together.md

@@ -37,17 +37,17 @@ The following environment variables can be configured:
 
 The following models are available by default:
 
-- `meta-llama/Llama-3.1-8B-Instruct`
+- `meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo (aliases: meta-llama/Llama-3.1-8B-Instruct)`
-- `meta-llama/Llama-3.1-70B-Instruct`
+- `meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo (aliases: meta-llama/Llama-3.1-70B-Instruct)`
-- `meta-llama/Llama-3.1-405B-Instruct-FP8`
+- `meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo (aliases: meta-llama/Llama-3.1-405B-Instruct-FP8)`
-- `meta-llama/Llama-3.2-3B-Instruct`
+- `meta-llama/Llama-3.2-3B-Instruct-Turbo (aliases: meta-llama/Llama-3.2-3B-Instruct)`
-- `meta-llama/Llama-3.2-11B-Vision-Instruct`
+- `meta-llama/Llama-3.2-11B-Vision-Instruct-Turbo (aliases: meta-llama/Llama-3.2-11B-Vision-Instruct)`
-- `meta-llama/Llama-3.2-90B-Vision-Instruct`
+- `meta-llama/Llama-3.2-90B-Vision-Instruct-Turbo (aliases: meta-llama/Llama-3.2-90B-Vision-Instruct)`
-- `meta-llama/Llama-3.3-70B-Instruct`
+- `meta-llama/Llama-3.3-70B-Instruct-Turbo (aliases: meta-llama/Llama-3.3-70B-Instruct)`
-- `meta-llama/Llama-Guard-3-8B`
+- `meta-llama/Meta-Llama-Guard-3-8B (aliases: meta-llama/Llama-Guard-3-8B)`
-- `meta-llama/Llama-Guard-3-11B-Vision`
+- `meta-llama/Llama-Guard-3-11B-Vision-Turbo (aliases: meta-llama/Llama-Guard-3-11B-Vision)`
-- `togethercomputer/m2-bert-80M-8k-retrieval`
+- `togethercomputer/m2-bert-80M-8k-retrieval `
-- `togethercomputer/m2-bert-80M-32k-retrieval`
+- `togethercomputer/m2-bert-80M-32k-retrieval `
 
 
 ### Prerequisite: API Keys

llama_stack/providers/remote/inference/cerebras/cerebras.py

@@ -46,14 +46,14 @@ from llama_stack.providers.utils.inference.prompt_adapter import (
 )
 
 from .config import CerebrasImplConfig
-from .models import model_entries
+from .models import MODEL_ENTRIES
 
 
 class CerebrasInferenceAdapter(ModelRegistryHelper, Inference):
     def __init__(self, config: CerebrasImplConfig) -> None:
         ModelRegistryHelper.__init__(
             self,
-            model_entries=model_entries,
+            model_entries=MODEL_ENTRIES,
         )
         self.config = config
 

llama_stack/providers/remote/inference/cerebras/models.py

@@ -9,7 +9,7 @@ from llama_stack.providers.utils.inference.model_registry import (
     build_hf_repo_model_entry,
 )
 
-model_entries = [
+MODEL_ENTRIES = [
     build_hf_repo_model_entry(
         "llama3.1-8b",
         CoreModelId.llama3_1_8b_instruct.value,

llama_stack/providers/remote/inference/groq/models.py

@@ -5,10 +5,13 @@
 # the root directory of this source tree.
 
 from llama_stack.models.llama.sku_list import CoreModelId
-from llama_stack.providers.utils.inference.model_registry import build_model_entry
+from llama_stack.providers.utils.inference.model_registry import (
+    build_hf_repo_model_entry,
+    build_model_entry,
+)
 
 MODEL_ENTRIES = [
-    build_model_entry(
+    build_hf_repo_model_entry(
         "groq/llama3-8b-8192",
         CoreModelId.llama3_1_8b_instruct.value,
     ),
@@ -16,11 +19,11 @@ MODEL_ENTRIES = [
         "groq/llama-3.1-8b-instant",
         CoreModelId.llama3_1_8b_instruct.value,
     ),
-    build_model_entry(
+    build_hf_repo_model_entry(
         "groq/llama3-70b-8192",
         CoreModelId.llama3_70b_instruct.value,
     ),
-    build_model_entry(
+    build_hf_repo_model_entry(
         "groq/llama-3.3-70b-versatile",
         CoreModelId.llama3_3_70b_instruct.value,
     ),
@@ -28,7 +31,7 @@ MODEL_ENTRIES = [
     # Preview models aren't recommended for production use, but we include this one
     # to pass the test fixture
     # TODO(aidand): Replace this with a stable model once Groq supports it
-    build_model_entry(
+    build_hf_repo_model_entry(
         "groq/llama-3.2-3b-preview",
         CoreModelId.llama3_2_3b_instruct.value,
     ),

llama_stack/providers/remote/inference/nvidia/models.py

@@ -11,7 +11,7 @@ from llama_stack.providers.utils.inference.model_registry import (
     build_hf_repo_model_entry,
 )
 
-_MODEL_ENTRIES = [
+MODEL_ENTRIES = [
     build_hf_repo_model_entry(
         "meta/llama3-8b-instruct",
         CoreModelId.llama3_8b_instruct.value,

llama_stack/providers/remote/inference/nvidia/nvidia.py

@@ -47,7 +47,7 @@ from llama_stack.providers.utils.inference.openai_compat import (
 from llama_stack.providers.utils.inference.prompt_adapter import content_has_media
 
 from . import NVIDIAConfig
-from .models import _MODEL_ENTRIES
+from .models import MODEL_ENTRIES
 from .openai_utils import (
     convert_chat_completion_request,
     convert_completion_request,
@@ -62,7 +62,7 @@ logger = logging.getLogger(__name__)
 class NVIDIAInferenceAdapter(Inference, ModelRegistryHelper):
     def __init__(self, config: NVIDIAConfig) -> None:
         # TODO(mf): filter by available models
-        ModelRegistryHelper.__init__(self, model_entries=_MODEL_ENTRIES)
+        ModelRegistryHelper.__init__(self, model_entries=MODEL_ENTRIES)
 
         logger.info(f"Initializing NVIDIAInferenceAdapter({config.url})...")
 

llama_stack/templates/bedrock/bedrock.py

@@ -6,12 +6,10 @@
 
 from pathlib import Path
 
-from llama_stack.apis.models import ModelInput
 from llama_stack.distribution.datatypes import Provider, ToolGroupInput
-from llama_stack.models.llama.sku_list import all_registered_models
 from llama_stack.providers.inline.vector_io.faiss.config import FaissVectorIOConfig
 from llama_stack.providers.remote.inference.bedrock.models import MODEL_ENTRIES
-from llama_stack.templates.template import DistributionTemplate, RunConfigSettings
+from llama_stack.templates.template import DistributionTemplate, RunConfigSettings, get_model_registry
 
 
 def get_distribution_template() -> DistributionTemplate:
@@ -39,16 +37,11 @@ def get_distribution_template() -> DistributionTemplate:
         config=FaissVectorIOConfig.sample_run_config(f"distributions/{name}"),
     )
 
-    core_model_to_hf_repo = {m.descriptor(): m.huggingface_repo for m in all_registered_models()}
+    available_models = {
+        "bedrock": MODEL_ENTRIES,
+    }
+    default_models = get_model_registry(available_models)
 
-    default_models = [
-        ModelInput(
-            model_id=core_model_to_hf_repo[m.llama_model],
-            provider_model_id=m.provider_model_id,
-            provider_id="bedrock",
-        )
-        for m in MODEL_ENTRIES
-    ]
     default_tool_groups = [
         ToolGroupInput(
             toolgroup_id="builtin::websearch",
@@ -71,7 +64,7 @@ def get_distribution_template() -> DistributionTemplate:
         container_image=None,
         template_path=Path(__file__).parent / "doc_template.md",
         providers=providers,
-        default_models=default_models,
+        available_models_by_provider=available_models,
         run_configs={
             "run.yaml": RunConfigSettings(
                 provider_overrides={

llama_stack/templates/bedrock/doc_template.md

@@ -28,7 +28,7 @@ The following environment variables can be configured:
 The following models are available by default:
 
 {% for model in default_models %}
-- `{{ model.model_id }} ({{ model.provider_model_id }})`
+- `{{ model.model_id }} {{ model.doc_string }}`
 {% endfor %}
 {% endif %}
 

llama_stack/templates/bedrock/run.yaml

@@ -88,16 +88,31 @@ metadata_store:
   type: sqlite
   db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/bedrock}/registry.db
 models:
+- metadata: {}
+  model_id: meta.llama3-1-8b-instruct-v1:0
+  provider_id: bedrock
+  provider_model_id: meta.llama3-1-8b-instruct-v1:0
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-8B-Instruct
   provider_id: bedrock
   provider_model_id: meta.llama3-1-8b-instruct-v1:0
   model_type: llm
+- metadata: {}
+  model_id: meta.llama3-1-70b-instruct-v1:0
+  provider_id: bedrock
+  provider_model_id: meta.llama3-1-70b-instruct-v1:0
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-70B-Instruct
   provider_id: bedrock
   provider_model_id: meta.llama3-1-70b-instruct-v1:0
   model_type: llm
+- metadata: {}
+  model_id: meta.llama3-1-405b-instruct-v1:0
+  provider_id: bedrock
+  provider_model_id: meta.llama3-1-405b-instruct-v1:0
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-405B-Instruct-FP8
   provider_id: bedrock

llama_stack/templates/cerebras/cerebras.py

@@ -8,14 +8,13 @@ from pathlib import Path
 
 from llama_stack.apis.models.models import ModelType
 from llama_stack.distribution.datatypes import ModelInput, Provider, ToolGroupInput
-from llama_stack.models.llama.sku_list import all_registered_models
 from llama_stack.providers.inline.inference.sentence_transformers import (
     SentenceTransformersInferenceConfig,
 )
 from llama_stack.providers.inline.vector_io.faiss.config import FaissVectorIOConfig
 from llama_stack.providers.remote.inference.cerebras import CerebrasImplConfig
-from llama_stack.providers.remote.inference.cerebras.models import model_entries
+from llama_stack.providers.remote.inference.cerebras.models import MODEL_ENTRIES
-from llama_stack.templates.template import DistributionTemplate, RunConfigSettings
+from llama_stack.templates.template import DistributionTemplate, RunConfigSettings, get_model_registry
 
 
 def get_distribution_template() -> DistributionTemplate:
@@ -48,15 +47,10 @@ def get_distribution_template() -> DistributionTemplate:
         config=SentenceTransformersInferenceConfig.sample_run_config(),
     )
 
-    core_model_to_hf_repo = {m.descriptor(): m.huggingface_repo for m in all_registered_models()}
+    available_models = {
-    default_models = [
+        "cerebras": MODEL_ENTRIES,
-        ModelInput(
+    }
-            model_id=core_model_to_hf_repo[m.llama_model],
+    default_models = get_model_registry(available_models)
-            provider_model_id=m.provider_model_id,
-            provider_id="cerebras",
-        )
-        for m in model_entries
-    ]
     embedding_model = ModelInput(
         model_id="all-MiniLM-L6-v2",
         provider_id="sentence-transformers",
@@ -92,7 +86,7 @@ def get_distribution_template() -> DistributionTemplate:
         container_image=None,
         template_path=Path(__file__).parent / "doc_template.md",
         providers=providers,
-        default_models=default_models,
+        available_models_by_provider=available_models,
         run_configs={
             "run.yaml": RunConfigSettings(
                 provider_overrides={

llama_stack/templates/cerebras/doc_template.md

@@ -20,7 +20,7 @@ The following environment variables can be configured:
 The following models are available by default:
 
 {% for model in default_models %}
-- `{{ model.model_id }} ({{ model.provider_model_id }})`
+- `{{ model.model_id }} {{ model.doc_string }}`
 {% endfor %}
 {% endif %}
 

llama_stack/templates/cerebras/run.yaml

@@ -90,11 +90,21 @@ metadata_store:
   type: sqlite
   db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/cerebras}/registry.db
 models:
+- metadata: {}
+  model_id: llama3.1-8b
+  provider_id: cerebras
+  provider_model_id: llama3.1-8b
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-8B-Instruct
   provider_id: cerebras
   provider_model_id: llama3.1-8b
   model_type: llm
+- metadata: {}
+  model_id: llama-3.3-70b
+  provider_id: cerebras
+  provider_model_id: llama-3.3-70b
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.3-70B-Instruct
   provider_id: cerebras

llama_stack/templates/ci-tests/ci_tests.py

@@ -12,14 +12,13 @@ from llama_stack.distribution.datatypes import (
     ShieldInput,
     ToolGroupInput,
 )
-from llama_stack.models.llama.sku_list import all_registered_models
 from llama_stack.providers.inline.inference.sentence_transformers import (
     SentenceTransformersInferenceConfig,
 )
 from llama_stack.providers.inline.vector_io.sqlite_vec.config import SQLiteVectorIOConfig
 from llama_stack.providers.remote.inference.fireworks.config import FireworksImplConfig
 from llama_stack.providers.remote.inference.fireworks.models import MODEL_ENTRIES
-from llama_stack.templates.template import DistributionTemplate, RunConfigSettings
+from llama_stack.templates.template import DistributionTemplate, RunConfigSettings, get_model_registry
 
 
 def get_distribution_template() -> DistributionTemplate:
@@ -71,16 +70,10 @@ def get_distribution_template() -> DistributionTemplate:
             provider_id="code-interpreter",
         ),
     ]
-    core_model_to_hf_repo = {m.descriptor(): m.huggingface_repo for m in all_registered_models()}
+    available_models = {
-    default_models = [
+        "fireworks": MODEL_ENTRIES,
-        ModelInput(
+    }
-            model_id=core_model_to_hf_repo[m.llama_model] if m.llama_model else m.provider_model_id,
+    default_models = get_model_registry(available_models)
-            provider_id="fireworks",
-            model_type=m.model_type,
-            metadata=m.metadata,
-        )
-        for m in MODEL_ENTRIES
-    ]
     embedding_model = ModelInput(
         model_id="all-MiniLM-L6-v2",
         provider_id="sentence-transformers",
@@ -97,7 +90,7 @@ def get_distribution_template() -> DistributionTemplate:
         container_image=None,
         template_path=None,
         providers=providers,
-        default_models=default_models + [embedding_model],
+        available_models_by_provider=available_models,
         run_configs={
             "run.yaml": RunConfigSettings(
                 provider_overrides={

llama_stack/templates/ci-tests/run.yaml

@@ -90,51 +90,112 @@ metadata_store:
   type: sqlite
   db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/ci-tests}/registry.db
 models:
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p1-8b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p1-8b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-8B-Instruct
   provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p1-8b-instruct
+  model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p1-70b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p1-70b-instruct
   model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-70B-Instruct
   provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p1-70b-instruct
+  model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p1-405b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p1-405b-instruct
   model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-405B-Instruct-FP8
   provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p1-405b-instruct
+  model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p2-1b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p2-1b-instruct
   model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-1B-Instruct
   provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p2-1b-instruct
+  model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p2-3b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p2-3b-instruct
   model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-3B-Instruct
   provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p2-3b-instruct
+  model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p2-11b-vision-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p2-11b-vision-instruct
   model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-11B-Vision-Instruct
   provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p2-11b-vision-instruct
+  model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p2-90b-vision-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p2-90b-vision-instruct
   model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-90B-Vision-Instruct
   provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p2-90b-vision-instruct
+  model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p3-70b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p3-70b-instruct
   model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.3-70B-Instruct
   provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p3-70b-instruct
+  model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-guard-3-8b
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-guard-3-8b
   model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-Guard-3-8B
   provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-guard-3-8b
+  model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-guard-3-11b-vision
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-guard-3-11b-vision
   model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-Guard-3-11B-Vision
   provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-guard-3-11b-vision
   model_type: llm
 - metadata:
     embedding_dimension: 768
     context_length: 8192
   model_id: nomic-ai/nomic-embed-text-v1.5
   provider_id: fireworks
+  provider_model_id: nomic-ai/nomic-embed-text-v1.5
   model_type: embedding
 - metadata:
     embedding_dimension: 384

llama_stack/templates/dell/dell.py

@@ -3,7 +3,6 @@
 #
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
-from pathlib import Path
 
 from llama_stack.apis.models.models import ModelType
 from llama_stack.distribution.datatypes import (
@@ -99,9 +98,7 @@ def get_distribution_template() -> DistributionTemplate:
         distro_type="self_hosted",
         description="Dell's distribution of Llama Stack. TGI inference via Dell's custom container",
         container_image=None,
-        template_path=Path(__file__).parent / "doc_template.md",
         providers=providers,
-        default_models=[inference_model, embedding_model],
         run_configs={
             "run.yaml": RunConfigSettings(
                 provider_overrides={

llama_stack/templates/dev/dev.py

@@ -13,7 +13,6 @@ from llama_stack.distribution.datatypes import (
     ShieldInput,
     ToolGroupInput,
 )
-from llama_stack.models.llama.sku_list import all_registered_models
 from llama_stack.providers.inline.inference.sentence_transformers import (
     SentenceTransformersInferenceConfig,
 )
@@ -28,7 +27,7 @@ from llama_stack.providers.remote.inference.groq.config import GroqConfig
 from llama_stack.providers.remote.inference.groq.models import MODEL_ENTRIES as GROQ_MODEL_ENTRIES
 from llama_stack.providers.remote.inference.openai.config import OpenAIConfig
 from llama_stack.providers.remote.inference.openai.models import MODEL_ENTRIES as OPENAI_MODEL_ENTRIES
-from llama_stack.templates.template import DistributionTemplate, RunConfigSettings
+from llama_stack.templates.template import DistributionTemplate, RunConfigSettings, get_model_registry
 
 
 def get_inference_providers() -> Tuple[List[Provider], List[ModelInput]]:
@@ -61,8 +60,7 @@ def get_inference_providers() -> Tuple[List[Provider], List[ModelInput]]:
         ),
     ]
     inference_providers = []
-    default_models = []
+    available_models = {}
-    core_model_to_hf_repo = {m.descriptor(): m.huggingface_repo for m in all_registered_models()}
     for provider_id, model_entries, config in providers:
         inference_providers.append(
             Provider(
@@ -71,21 +69,12 @@ def get_inference_providers() -> Tuple[List[Provider], List[ModelInput]]:
                 config=config,
             )
         )
-        default_models.extend(
+        available_models[provider_id] = model_entries
-            ModelInput(
+    return inference_providers, available_models
-                model_id=core_model_to_hf_repo[m.llama_model] if m.llama_model else m.provider_model_id,
-                provider_model_id=m.provider_model_id,
-                provider_id=provider_id,
-                model_type=m.model_type,
-                metadata=m.metadata,
-            )
-            for m in model_entries
-        )
-    return inference_providers, default_models
 
 
 def get_distribution_template() -> DistributionTemplate:
-    inference_providers, default_models = get_inference_providers()
+    inference_providers, available_models = get_inference_providers()
     providers = {
         "inference": ([p.provider_type for p in inference_providers] + ["inline::sentence-transformers"]),
         "vector_io": ["inline::sqlite-vec", "remote::chromadb", "remote::pgvector"],
@@ -139,6 +128,7 @@ def get_distribution_template() -> DistributionTemplate:
         },
     )
 
+    default_models = get_model_registry(available_models)
     return DistributionTemplate(
         name=name,
         distro_type="self_hosted",
@@ -146,7 +136,7 @@ def get_distribution_template() -> DistributionTemplate:
         container_image=None,
         template_path=None,
         providers=providers,
-        default_models=[],
+        available_models_by_provider=available_models,
         run_configs={
             "run.yaml": RunConfigSettings(
                 provider_overrides={

llama_stack/templates/dev/run.yaml

@@ -136,51 +136,101 @@ models:
   provider_id: openai
   provider_model_id: openai/text-embedding-3-large
   model_type: embedding
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p1-8b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p1-8b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-8B-Instruct
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p1-8b-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p1-70b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p1-70b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-70B-Instruct
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p1-70b-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p1-405b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p1-405b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-405B-Instruct-FP8
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p1-405b-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p2-1b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p2-1b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-1B-Instruct
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p2-1b-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p2-3b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p2-3b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-3B-Instruct
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p2-3b-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p2-11b-vision-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p2-11b-vision-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-11B-Vision-Instruct
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p2-11b-vision-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p2-90b-vision-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p2-90b-vision-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-90B-Vision-Instruct
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p2-90b-vision-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p3-70b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p3-70b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.3-70B-Instruct
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p3-70b-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-guard-3-8b
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-guard-3-8b
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-Guard-3-8B
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-guard-3-8b
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-guard-3-11b-vision
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-guard-3-11b-vision
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-Guard-3-11B-Vision
   provider_id: fireworks
@@ -247,25 +297,45 @@ models:
   provider_model_id: gemini/text-embedding-004
   model_type: embedding
 - metadata: {}
-  model_id: meta-llama/Llama-3.1-8B-Instruct
+  model_id: groq/llama3-8b-8192
   provider_id: groq
   provider_model_id: groq/llama3-8b-8192
   model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-8B-Instruct
   provider_id: groq
+  provider_model_id: groq/llama3-8b-8192
+  model_type: llm
+- metadata: {}
+  model_id: groq/llama-3.1-8b-instant
+  provider_id: groq
   provider_model_id: groq/llama-3.1-8b-instant
   model_type: llm
+- metadata: {}
+  model_id: groq/llama3-70b-8192
+  provider_id: groq
+  provider_model_id: groq/llama3-70b-8192
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3-70B-Instruct
   provider_id: groq
   provider_model_id: groq/llama3-70b-8192
   model_type: llm
+- metadata: {}
+  model_id: groq/llama-3.3-70b-versatile
+  provider_id: groq
+  provider_model_id: groq/llama-3.3-70b-versatile
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.3-70B-Instruct
   provider_id: groq
   provider_model_id: groq/llama-3.3-70b-versatile
   model_type: llm
+- metadata: {}
+  model_id: groq/llama-3.2-3b-preview
+  provider_id: groq
+  provider_model_id: groq/llama-3.2-3b-preview
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-3B-Instruct
   provider_id: groq

llama_stack/templates/fireworks/doc_template.md

@@ -30,7 +30,7 @@ The following environment variables can be configured:
 The following models are available by default:
 
 {% for model in default_models %}
-- `{{ model.model_id }} ({{ model.provider_model_id }})`
+- `{{ model.model_id }} {{ model.doc_string }}`
 {% endfor %}
 {% endif %}
 

llama_stack/templates/fireworks/fireworks.py

@@ -13,14 +13,13 @@ from llama_stack.distribution.datatypes import (
     ShieldInput,
     ToolGroupInput,
 )
-from llama_stack.models.llama.sku_list import all_registered_models
 from llama_stack.providers.inline.inference.sentence_transformers import (
     SentenceTransformersInferenceConfig,
 )
 from llama_stack.providers.inline.vector_io.faiss.config import FaissVectorIOConfig
 from llama_stack.providers.remote.inference.fireworks.config import FireworksImplConfig
 from llama_stack.providers.remote.inference.fireworks.models import MODEL_ENTRIES
-from llama_stack.templates.template import DistributionTemplate, RunConfigSettings
+from llama_stack.templates.template import DistributionTemplate, RunConfigSettings, get_model_registry
 
 
 def get_distribution_template() -> DistributionTemplate:
@@ -60,17 +59,11 @@ def get_distribution_template() -> DistributionTemplate:
         config=FaissVectorIOConfig.sample_run_config(f"distributions/{name}"),
     )
 
-    core_model_to_hf_repo = {m.descriptor(): m.huggingface_repo for m in all_registered_models()}
+    available_models = {
-    default_models = [
+        "fireworks": MODEL_ENTRIES,
-        ModelInput(
+    }
-            model_id=core_model_to_hf_repo[m.llama_model] if m.llama_model else m.provider_model_id,
+    default_models = get_model_registry(available_models)
-            provider_model_id=m.provider_model_id,
+
-            provider_id="fireworks",
-            metadata=m.metadata,
-            model_type=m.model_type,
-        )
-        for m in MODEL_ENTRIES
-    ]
     embedding_model = ModelInput(
         model_id="all-MiniLM-L6-v2",
         provider_id="sentence-transformers",
@@ -101,7 +94,7 @@ def get_distribution_template() -> DistributionTemplate:
         container_image=None,
         template_path=Path(__file__).parent / "doc_template.md",
         providers=providers,
-        default_models=default_models,
+        available_models_by_provider=available_models,
         run_configs={
             "run.yaml": RunConfigSettings(
                 provider_overrides={

llama_stack/templates/fireworks/run-with-safety.yaml

@@ -99,51 +99,101 @@ metadata_store:
   type: sqlite
   db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/fireworks}/registry.db
 models:
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p1-8b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p1-8b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-8B-Instruct
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p1-8b-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p1-70b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p1-70b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-70B-Instruct
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p1-70b-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p1-405b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p1-405b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-405B-Instruct-FP8
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p1-405b-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p2-1b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p2-1b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-1B-Instruct
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p2-1b-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p2-3b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p2-3b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-3B-Instruct
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p2-3b-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p2-11b-vision-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p2-11b-vision-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-11B-Vision-Instruct
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p2-11b-vision-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p2-90b-vision-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p2-90b-vision-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-90B-Vision-Instruct
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p2-90b-vision-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p3-70b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p3-70b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.3-70B-Instruct
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p3-70b-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-guard-3-8b
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-guard-3-8b
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-Guard-3-8B
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-guard-3-8b
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-guard-3-11b-vision
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-guard-3-11b-vision
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-Guard-3-11B-Vision
   provider_id: fireworks

llama_stack/templates/fireworks/run.yaml

@@ -93,51 +93,101 @@ metadata_store:
   type: sqlite
   db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/fireworks}/registry.db
 models:
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p1-8b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p1-8b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-8B-Instruct
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p1-8b-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p1-70b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p1-70b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-70B-Instruct
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p1-70b-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p1-405b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p1-405b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-405B-Instruct-FP8
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p1-405b-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p2-1b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p2-1b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-1B-Instruct
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p2-1b-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p2-3b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p2-3b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-3B-Instruct
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p2-3b-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p2-11b-vision-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p2-11b-vision-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-11B-Vision-Instruct
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p2-11b-vision-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p2-90b-vision-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p2-90b-vision-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-90B-Vision-Instruct
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p2-90b-vision-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-v3p3-70b-instruct
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-v3p3-70b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.3-70B-Instruct
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-v3p3-70b-instruct
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-guard-3-8b
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-guard-3-8b
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-Guard-3-8B
   provider_id: fireworks
   provider_model_id: accounts/fireworks/models/llama-guard-3-8b
   model_type: llm
+- metadata: {}
+  model_id: accounts/fireworks/models/llama-guard-3-11b-vision
+  provider_id: fireworks
+  provider_model_id: accounts/fireworks/models/llama-guard-3-11b-vision
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-Guard-3-11B-Vision
   provider_id: fireworks

llama_stack/templates/groq/doc_template.md

@@ -30,7 +30,7 @@ The following environment variables can be configured:
 The following models are available by default:
 
 {% for model in default_models %}
-- `{{ model.model_id }} ({{ model.provider_model_id }})`
+- `{{ model.model_id }} {{ model.doc_string }}`
 {% endfor %}
 {% endif %}
 

llama_stack/templates/groq/groq.py

@@ -12,13 +12,12 @@ from llama_stack.distribution.datatypes import (
     Provider,
     ToolGroupInput,
 )
-from llama_stack.models.llama.sku_list import all_registered_models
 from llama_stack.providers.inline.inference.sentence_transformers import (
     SentenceTransformersInferenceConfig,
 )
 from llama_stack.providers.remote.inference.groq import GroqConfig
 from llama_stack.providers.remote.inference.groq.models import MODEL_ENTRIES
-from llama_stack.templates.template import DistributionTemplate, RunConfigSettings
+from llama_stack.templates.template import DistributionTemplate, RunConfigSettings, get_model_registry
 
 
 def get_distribution_template() -> DistributionTemplate:
@@ -60,18 +59,10 @@ def get_distribution_template() -> DistributionTemplate:
         },
     )
 
-    core_model_to_hf_repo = {m.descriptor(): m.huggingface_repo for m in all_registered_models()}
+    available_models = {
-    default_models = [
+        "groq": MODEL_ENTRIES,
-        ModelInput(
+    }
-            model_id=core_model_to_hf_repo[m.llama_model] if m.llama_model else m.provider_model_id,
+    default_models = get_model_registry(available_models)
-            provider_model_id=m.provider_model_id,
-            provider_id=name,
-            model_type=m.model_type,
-            metadata=m.metadata,
-        )
-        for m in MODEL_ENTRIES
-    ]
-
     default_tool_groups = [
         ToolGroupInput(
             toolgroup_id="builtin::websearch",
@@ -94,7 +85,7 @@ def get_distribution_template() -> DistributionTemplate:
         docker_image=None,
         template_path=Path(__file__).parent / "doc_template.md",
         providers=providers,
-        default_models=default_models,
+        available_models_by_provider=available_models,
         run_configs={
             "run.yaml": RunConfigSettings(
                 provider_overrides={

llama_stack/templates/groq/run.yaml

@@ -91,25 +91,45 @@ metadata_store:
   db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/groq}/registry.db
 models:
 - metadata: {}
-  model_id: meta-llama/Llama-3.1-8B-Instruct
+  model_id: groq/llama3-8b-8192
   provider_id: groq
   provider_model_id: groq/llama3-8b-8192
   model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-8B-Instruct
   provider_id: groq
+  provider_model_id: groq/llama3-8b-8192
+  model_type: llm
+- metadata: {}
+  model_id: groq/llama-3.1-8b-instant
+  provider_id: groq
   provider_model_id: groq/llama-3.1-8b-instant
   model_type: llm
+- metadata: {}
+  model_id: groq/llama3-70b-8192
+  provider_id: groq
+  provider_model_id: groq/llama3-70b-8192
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3-70B-Instruct
   provider_id: groq
   provider_model_id: groq/llama3-70b-8192
   model_type: llm
+- metadata: {}
+  model_id: groq/llama-3.3-70b-versatile
+  provider_id: groq
+  provider_model_id: groq/llama-3.3-70b-versatile
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.3-70B-Instruct
   provider_id: groq
   provider_model_id: groq/llama-3.3-70b-versatile
   model_type: llm
+- metadata: {}
+  model_id: groq/llama-3.2-3b-preview
+  provider_id: groq
+  provider_model_id: groq/llama-3.2-3b-preview
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-3B-Instruct
   provider_id: groq

llama_stack/templates/hf-endpoint/hf_endpoint.py

@@ -92,7 +92,6 @@ def get_distribution_template() -> DistributionTemplate:
         container_image=None,
         template_path=None,
         providers=providers,
-        default_models=[inference_model, safety_model],
         run_configs={
             "run.yaml": RunConfigSettings(
                 provider_overrides={

llama_stack/templates/hf-serverless/hf_serverless.py

@@ -93,7 +93,6 @@ def get_distribution_template() -> DistributionTemplate:
         container_image=None,
         template_path=None,
         providers=providers,
-        default_models=[inference_model, safety_model],
         run_configs={
             "run.yaml": RunConfigSettings(
                 provider_overrides={

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

@@ -98,7 +98,6 @@ def get_distribution_template() -> DistributionTemplate:
         description="Use Meta Reference for running LLM inference",
         template_path=Path(__file__).parent / "doc_template.md",
         providers=providers,
-        default_models=[inference_model, safety_model],
         run_configs={
             "run.yaml": RunConfigSettings(
                 provider_overrides={

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

@@ -88,7 +88,6 @@ def get_distribution_template() -> DistributionTemplate:
         description="Use Meta Reference with fp8, int4 quantization for running LLM inference",
         template_path=Path(__file__).parent / "doc_template.md",
         providers=providers,
-        default_models=[inference_model],
         run_configs={
             "run.yaml": RunConfigSettings(
                 provider_overrides={

llama_stack/templates/nvidia/doc_template.md

@@ -20,7 +20,7 @@ The following environment variables can be configured:
 The following models are available by default:
 
 {% for model in default_models %}
-- `{{ model.model_id }} ({{ model.provider_model_id }})`
+- `{{ model.model_id }} {{ model.doc_string }}`
 {% endfor %}
 {% endif %}
 

llama_stack/templates/nvidia/nvidia.py

@@ -6,11 +6,10 @@
 
 from pathlib import Path
 
-from llama_stack.distribution.datatypes import ModelInput, Provider, ToolGroupInput
+from llama_stack.distribution.datatypes import Provider, ToolGroupInput
-from llama_stack.models.llama.sku_list import all_registered_models
 from llama_stack.providers.remote.inference.nvidia import NVIDIAConfig
-from llama_stack.providers.remote.inference.nvidia.models import _MODEL_ENTRIES
+from llama_stack.providers.remote.inference.nvidia.models import MODEL_ENTRIES
-from llama_stack.templates.template import DistributionTemplate, RunConfigSettings
+from llama_stack.templates.template import DistributionTemplate, RunConfigSettings, get_model_registry
 
 
 def get_distribution_template() -> DistributionTemplate:
@@ -38,17 +37,9 @@ def get_distribution_template() -> DistributionTemplate:
         config=NVIDIAConfig.sample_run_config(),
     )
 
-    core_model_to_hf_repo = {m.descriptor(): m.huggingface_repo for m in all_registered_models()}
+    available_models = {
-    default_models = [
+        "nvidia": MODEL_ENTRIES,
-        ModelInput(
+    }
-            model_id=core_model_to_hf_repo[m.llama_model] if m.llama_model else m.provider_model_id,
-            provider_model_id=m.provider_model_id,
-            provider_id="nvidia",
-            model_type=m.model_type,
-            metadata=m.metadata,
-        )
-        for m in _MODEL_ENTRIES
-    ]
     default_tool_groups = [
         ToolGroupInput(
             toolgroup_id="builtin::websearch",
@@ -64,6 +55,7 @@ def get_distribution_template() -> DistributionTemplate:
         ),
     ]
 
+    default_models = get_model_registry(available_models)
     return DistributionTemplate(
         name="nvidia",
         distro_type="remote_hosted",
@@ -71,7 +63,7 @@ def get_distribution_template() -> DistributionTemplate:
         container_image=None,
         template_path=Path(__file__).parent / "doc_template.md",
         providers=providers,
-        default_models=default_models,
+        available_models_by_provider=available_models,
         run_configs={
             "run.yaml": RunConfigSettings(
                 provider_overrides={

llama_stack/templates/nvidia/run.yaml

@@ -90,46 +90,91 @@ metadata_store:
   type: sqlite
   db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/nvidia}/registry.db
 models:
+- metadata: {}
+  model_id: meta/llama3-8b-instruct
+  provider_id: nvidia
+  provider_model_id: meta/llama3-8b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3-8B-Instruct
   provider_id: nvidia
   provider_model_id: meta/llama3-8b-instruct
   model_type: llm
+- metadata: {}
+  model_id: meta/llama3-70b-instruct
+  provider_id: nvidia
+  provider_model_id: meta/llama3-70b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3-70B-Instruct
   provider_id: nvidia
   provider_model_id: meta/llama3-70b-instruct
   model_type: llm
+- metadata: {}
+  model_id: meta/llama-3.1-8b-instruct
+  provider_id: nvidia
+  provider_model_id: meta/llama-3.1-8b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-8B-Instruct
   provider_id: nvidia
   provider_model_id: meta/llama-3.1-8b-instruct
   model_type: llm
+- metadata: {}
+  model_id: meta/llama-3.1-70b-instruct
+  provider_id: nvidia
+  provider_model_id: meta/llama-3.1-70b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-70B-Instruct
   provider_id: nvidia
   provider_model_id: meta/llama-3.1-70b-instruct
   model_type: llm
+- metadata: {}
+  model_id: meta/llama-3.1-405b-instruct
+  provider_id: nvidia
+  provider_model_id: meta/llama-3.1-405b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-405B-Instruct-FP8
   provider_id: nvidia
   provider_model_id: meta/llama-3.1-405b-instruct
   model_type: llm
+- metadata: {}
+  model_id: meta/llama-3.2-1b-instruct
+  provider_id: nvidia
+  provider_model_id: meta/llama-3.2-1b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-1B-Instruct
   provider_id: nvidia
   provider_model_id: meta/llama-3.2-1b-instruct
   model_type: llm
+- metadata: {}
+  model_id: meta/llama-3.2-3b-instruct
+  provider_id: nvidia
+  provider_model_id: meta/llama-3.2-3b-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-3B-Instruct
   provider_id: nvidia
   provider_model_id: meta/llama-3.2-3b-instruct
   model_type: llm
+- metadata: {}
+  model_id: meta/llama-3.2-11b-vision-instruct
+  provider_id: nvidia
+  provider_model_id: meta/llama-3.2-11b-vision-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-11B-Vision-Instruct
   provider_id: nvidia
   provider_model_id: meta/llama-3.2-11b-vision-instruct
   model_type: llm
+- metadata: {}
+  model_id: meta/llama-3.2-90b-vision-instruct
+  provider_id: nvidia
+  provider_model_id: meta/llama-3.2-90b-vision-instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-90B-Vision-Instruct
   provider_id: nvidia

llama_stack/templates/ollama/ollama.py

@@ -87,7 +87,6 @@ def get_distribution_template() -> DistributionTemplate:
         container_image=None,
         template_path=Path(__file__).parent / "doc_template.md",
         providers=providers,
-        default_models=[inference_model, safety_model],
         run_configs={
             "run.yaml": RunConfigSettings(
                 provider_overrides={

llama_stack/templates/remote-vllm/vllm.py

@@ -95,7 +95,6 @@ def get_distribution_template() -> DistributionTemplate:
         description="Use (an external) vLLM server for running LLM inference",
         template_path=Path(__file__).parent / "doc_template.md",
         providers=providers,
-        default_models=[inference_model, safety_model],
         run_configs={
             "run.yaml": RunConfigSettings(
                 provider_overrides={

llama_stack/templates/sambanova/doc_template.md

@@ -30,7 +30,7 @@ The following environment variables can be configured:
 The following models are available by default:
 
 {% for model in default_models %}
-- `{{ model.model_id }} ({{ model.provider_model_id }})`
+- `{{ model.model_id }} {{ model.doc_string }}`
 {% endfor %}
 {% endif %}
 

llama_stack/templates/sambanova/run.yaml

@@ -68,46 +68,91 @@ metadata_store:
   type: sqlite
   db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/sambanova}/registry.db
 models:
+- metadata: {}
+  model_id: Meta-Llama-3.1-8B-Instruct
+  provider_id: sambanova
+  provider_model_id: Meta-Llama-3.1-8B-Instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-8B-Instruct
   provider_id: sambanova
   provider_model_id: Meta-Llama-3.1-8B-Instruct
   model_type: llm
+- metadata: {}
+  model_id: Meta-Llama-3.1-70B-Instruct
+  provider_id: sambanova
+  provider_model_id: Meta-Llama-3.1-70B-Instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-70B-Instruct
   provider_id: sambanova
   provider_model_id: Meta-Llama-3.1-70B-Instruct
   model_type: llm
+- metadata: {}
+  model_id: Meta-Llama-3.1-405B-Instruct
+  provider_id: sambanova
+  provider_model_id: Meta-Llama-3.1-405B-Instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-405B-Instruct-FP8
   provider_id: sambanova
   provider_model_id: Meta-Llama-3.1-405B-Instruct
   model_type: llm
+- metadata: {}
+  model_id: Meta-Llama-3.2-1B-Instruct
+  provider_id: sambanova
+  provider_model_id: Meta-Llama-3.2-1B-Instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-1B-Instruct
   provider_id: sambanova
   provider_model_id: Meta-Llama-3.2-1B-Instruct
   model_type: llm
+- metadata: {}
+  model_id: Meta-Llama-3.2-3B-Instruct
+  provider_id: sambanova
+  provider_model_id: Meta-Llama-3.2-3B-Instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-3B-Instruct
   provider_id: sambanova
   provider_model_id: Meta-Llama-3.2-3B-Instruct
   model_type: llm
+- metadata: {}
+  model_id: Meta-Llama-3.3-70B-Instruct
+  provider_id: sambanova
+  provider_model_id: Meta-Llama-3.3-70B-Instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.3-70B-Instruct
   provider_id: sambanova
   provider_model_id: Meta-Llama-3.3-70B-Instruct
   model_type: llm
+- metadata: {}
+  model_id: Llama-3.2-11B-Vision-Instruct
+  provider_id: sambanova
+  provider_model_id: Llama-3.2-11B-Vision-Instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-11B-Vision-Instruct
   provider_id: sambanova
   provider_model_id: Llama-3.2-11B-Vision-Instruct
   model_type: llm
+- metadata: {}
+  model_id: Llama-3.2-90B-Vision-Instruct
+  provider_id: sambanova
+  provider_model_id: Llama-3.2-90B-Vision-Instruct
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-90B-Vision-Instruct
   provider_id: sambanova
   provider_model_id: Llama-3.2-90B-Vision-Instruct
   model_type: llm
+- metadata: {}
+  model_id: Meta-Llama-Guard-3-8B
+  provider_id: sambanova
+  provider_model_id: Meta-Llama-Guard-3-8B
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-Guard-3-8B
   provider_id: sambanova

llama_stack/templates/sambanova/sambanova.py

@@ -7,15 +7,13 @@
 from pathlib import Path
 
 from llama_stack.distribution.datatypes import (
-    ModelInput,
     Provider,
     ShieldInput,
     ToolGroupInput,
 )
-from llama_stack.models.llama.sku_list import all_registered_models
 from llama_stack.providers.remote.inference.sambanova import SambaNovaImplConfig
 from llama_stack.providers.remote.inference.sambanova.models import MODEL_ENTRIES
-from llama_stack.templates.template import DistributionTemplate, RunConfigSettings
+from llama_stack.templates.template import DistributionTemplate, RunConfigSettings, get_model_registry
 
 
 def get_distribution_template() -> DistributionTemplate:
@@ -40,16 +38,10 @@ def get_distribution_template() -> DistributionTemplate:
         config=SambaNovaImplConfig.sample_run_config(),
     )
 
-    core_model_to_hf_repo = {m.descriptor(): m.huggingface_repo for m in all_registered_models()}
+    available_models = {
-    default_models = [
+        name: MODEL_ENTRIES,
-        ModelInput(
+    }
-            model_id=core_model_to_hf_repo[m.llama_model],
+    default_models = get_model_registry(available_models)
-            provider_model_id=m.provider_model_id,
-            provider_id=name,
-        )
-        for m in MODEL_ENTRIES
-    ]
-
     default_tool_groups = [
         ToolGroupInput(
             toolgroup_id="builtin::websearch",
@@ -72,7 +64,7 @@ def get_distribution_template() -> DistributionTemplate:
         docker_image=None,
         template_path=Path(__file__).parent / "doc_template.md",
         providers=providers,
-        default_models=default_models,
+        available_models_by_provider=available_models,
         run_configs={
             "run.yaml": RunConfigSettings(
                 provider_overrides={

llama_stack/templates/template.py

@@ -24,9 +24,33 @@ from llama_stack.distribution.datatypes import (
 )
 from llama_stack.distribution.distribution import get_provider_registry
 from llama_stack.distribution.utils.dynamic import instantiate_class_type
+from llama_stack.providers.utils.inference.model_registry import ProviderModelEntry
 from llama_stack.providers.utils.kvstore.config import SqliteKVStoreConfig
 
 
+def get_model_registry(available_models: Dict[str, List[ProviderModelEntry]]) -> List[ModelInput]:
+    models = []
+    for provider_id, entries in available_models.items():
+        for entry in entries:
+            ids = [entry.provider_model_id] + entry.aliases
+            for model_id in ids:
+                models.append(
+                    ModelInput(
+                        model_id=model_id,
+                        provider_model_id=entry.provider_model_id,
+                        provider_id=provider_id,
+                        model_type=entry.model_type,
+                        metadata=entry.metadata,
+                    )
+                )
+    return models
+
+
+class DefaultModel(BaseModel):
+    model_id: str
+    doc_string: str
+
+
 class RunConfigSettings(BaseModel):
     provider_overrides: Dict[str, List[Provider]] = Field(default_factory=dict)
     default_models: Optional[List[ModelInput]] = None
@@ -110,7 +134,7 @@ class DistributionTemplate(BaseModel):
     run_config_env_vars: Optional[Dict[str, Tuple[str, str]]] = None
     container_image: Optional[str] = None
 
-    default_models: Optional[List[ModelInput]] = None
+    available_models_by_provider: Optional[Dict[str, List[ProviderModelEntry]]] = None
 
     def build_config(self) -> BuildConfig:
         return BuildConfig(
@@ -148,13 +172,32 @@ class DistributionTemplate(BaseModel):
             autoescape=True,
         )
         template = env.from_string(template)
+
+        default_models = []
+        if self.available_models_by_provider:
+            has_multiple_providers = len(self.available_models_by_provider.keys()) > 1
+            for provider_id, model_entries in self.available_models_by_provider.items():
+                for model_entry in model_entries:
+                    doc_parts = []
+                    if model_entry.aliases:
+                        doc_parts.append(f"aliases: {', '.join(model_entry.aliases)}")
+                    if has_multiple_providers:
+                        doc_parts.append(f"provider: {provider_id}")
+
+                    default_models.append(
+                        DefaultModel(
+                            model_id=model_entry.provider_model_id,
+                            doc_string=f"({' -- '.join(doc_parts)})" if doc_parts else "",
+                        )
+                    )
+
         return template.render(
             name=self.name,
             description=self.description,
             providers=self.providers,
             providers_table=providers_table,
             run_config_env_vars=self.run_config_env_vars,
-            default_models=self.default_models,
+            default_models=default_models,
         )
 
     def save_distribution(self, yaml_output_dir: Path, doc_output_dir: Path) -> None:

llama_stack/templates/tgi/tgi.py

@@ -96,7 +96,6 @@ def get_distribution_template() -> DistributionTemplate:
         container_image=None,
         template_path=Path(__file__).parent / "doc_template.md",
         providers=providers,
-        default_models=[inference_model, safety_model],
         run_configs={
             "run.yaml": RunConfigSettings(
                 provider_overrides={

llama_stack/templates/together/doc_template.md

@@ -30,7 +30,7 @@ The following environment variables can be configured:
 The following models are available by default:
 
 {% for model in default_models %}
-- `{{ model.model_id }}`
+- `{{ model.model_id }} {{ model.doc_string }}`
 {% endfor %}
 {% endif %}
 

llama_stack/templates/together/run-with-safety.yaml

@@ -99,46 +99,91 @@ metadata_store:
   type: sqlite
   db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/together}/registry.db
 models:
+- metadata: {}
+  model_id: meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo
+  provider_id: together
+  provider_model_id: meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-8B-Instruct
   provider_id: together
   provider_model_id: meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo
   model_type: llm
+- metadata: {}
+  model_id: meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo
+  provider_id: together
+  provider_model_id: meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-70B-Instruct
   provider_id: together
   provider_model_id: meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo
   model_type: llm
+- metadata: {}
+  model_id: meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo
+  provider_id: together
+  provider_model_id: meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-405B-Instruct-FP8
   provider_id: together
   provider_model_id: meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo
   model_type: llm
+- metadata: {}
+  model_id: meta-llama/Llama-3.2-3B-Instruct-Turbo
+  provider_id: together
+  provider_model_id: meta-llama/Llama-3.2-3B-Instruct-Turbo
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-3B-Instruct
   provider_id: together
   provider_model_id: meta-llama/Llama-3.2-3B-Instruct-Turbo
   model_type: llm
+- metadata: {}
+  model_id: meta-llama/Llama-3.2-11B-Vision-Instruct-Turbo
+  provider_id: together
+  provider_model_id: meta-llama/Llama-3.2-11B-Vision-Instruct-Turbo
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-11B-Vision-Instruct
   provider_id: together
   provider_model_id: meta-llama/Llama-3.2-11B-Vision-Instruct-Turbo
   model_type: llm
+- metadata: {}
+  model_id: meta-llama/Llama-3.2-90B-Vision-Instruct-Turbo
+  provider_id: together
+  provider_model_id: meta-llama/Llama-3.2-90B-Vision-Instruct-Turbo
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-90B-Vision-Instruct
   provider_id: together
   provider_model_id: meta-llama/Llama-3.2-90B-Vision-Instruct-Turbo
   model_type: llm
+- metadata: {}
+  model_id: meta-llama/Llama-3.3-70B-Instruct-Turbo
+  provider_id: together
+  provider_model_id: meta-llama/Llama-3.3-70B-Instruct-Turbo
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.3-70B-Instruct
   provider_id: together
   provider_model_id: meta-llama/Llama-3.3-70B-Instruct-Turbo
   model_type: llm
+- metadata: {}
+  model_id: meta-llama/Meta-Llama-Guard-3-8B
+  provider_id: together
+  provider_model_id: meta-llama/Meta-Llama-Guard-3-8B
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-Guard-3-8B
   provider_id: together
   provider_model_id: meta-llama/Meta-Llama-Guard-3-8B
   model_type: llm
+- metadata: {}
+  model_id: meta-llama/Llama-Guard-3-11B-Vision-Turbo
+  provider_id: together
+  provider_model_id: meta-llama/Llama-Guard-3-11B-Vision-Turbo
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-Guard-3-11B-Vision
   provider_id: together

llama_stack/templates/together/run.yaml

@@ -93,46 +93,91 @@ metadata_store:
   type: sqlite
   db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/together}/registry.db
 models:
+- metadata: {}
+  model_id: meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo
+  provider_id: together
+  provider_model_id: meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-8B-Instruct
   provider_id: together
   provider_model_id: meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo
   model_type: llm
+- metadata: {}
+  model_id: meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo
+  provider_id: together
+  provider_model_id: meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-70B-Instruct
   provider_id: together
   provider_model_id: meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo
   model_type: llm
+- metadata: {}
+  model_id: meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo
+  provider_id: together
+  provider_model_id: meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.1-405B-Instruct-FP8
   provider_id: together
   provider_model_id: meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo
   model_type: llm
+- metadata: {}
+  model_id: meta-llama/Llama-3.2-3B-Instruct-Turbo
+  provider_id: together
+  provider_model_id: meta-llama/Llama-3.2-3B-Instruct-Turbo
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-3B-Instruct
   provider_id: together
   provider_model_id: meta-llama/Llama-3.2-3B-Instruct-Turbo
   model_type: llm
+- metadata: {}
+  model_id: meta-llama/Llama-3.2-11B-Vision-Instruct-Turbo
+  provider_id: together
+  provider_model_id: meta-llama/Llama-3.2-11B-Vision-Instruct-Turbo
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-11B-Vision-Instruct
   provider_id: together
   provider_model_id: meta-llama/Llama-3.2-11B-Vision-Instruct-Turbo
   model_type: llm
+- metadata: {}
+  model_id: meta-llama/Llama-3.2-90B-Vision-Instruct-Turbo
+  provider_id: together
+  provider_model_id: meta-llama/Llama-3.2-90B-Vision-Instruct-Turbo
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.2-90B-Vision-Instruct
   provider_id: together
   provider_model_id: meta-llama/Llama-3.2-90B-Vision-Instruct-Turbo
   model_type: llm
+- metadata: {}
+  model_id: meta-llama/Llama-3.3-70B-Instruct-Turbo
+  provider_id: together
+  provider_model_id: meta-llama/Llama-3.3-70B-Instruct-Turbo
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-3.3-70B-Instruct
   provider_id: together
   provider_model_id: meta-llama/Llama-3.3-70B-Instruct-Turbo
   model_type: llm
+- metadata: {}
+  model_id: meta-llama/Meta-Llama-Guard-3-8B
+  provider_id: together
+  provider_model_id: meta-llama/Meta-Llama-Guard-3-8B
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-Guard-3-8B
   provider_id: together
   provider_model_id: meta-llama/Meta-Llama-Guard-3-8B
   model_type: llm
+- metadata: {}
+  model_id: meta-llama/Llama-Guard-3-11B-Vision-Turbo
+  provider_id: together
+  provider_model_id: meta-llama/Llama-Guard-3-11B-Vision-Turbo
+  model_type: llm
 - metadata: {}
   model_id: meta-llama/Llama-Guard-3-11B-Vision
   provider_id: together

llama_stack/templates/together/together.py

@@ -13,14 +13,13 @@ from llama_stack.distribution.datatypes import (
     ShieldInput,
     ToolGroupInput,
 )
-from llama_stack.models.llama.sku_list import all_registered_models
 from llama_stack.providers.inline.inference.sentence_transformers import (
     SentenceTransformersInferenceConfig,
 )
 from llama_stack.providers.inline.vector_io.faiss.config import FaissVectorIOConfig
 from llama_stack.providers.remote.inference.together import TogetherImplConfig
 from llama_stack.providers.remote.inference.together.models import MODEL_ENTRIES
-from llama_stack.templates.template import DistributionTemplate, RunConfigSettings
+from llama_stack.templates.template import DistributionTemplate, RunConfigSettings, get_model_registry
 
 
 def get_distribution_template() -> DistributionTemplate:
@@ -57,18 +56,10 @@ def get_distribution_template() -> DistributionTemplate:
         provider_type="inline::sentence-transformers",
         config=SentenceTransformersInferenceConfig.sample_run_config(),
     )
-
+    available_models = {
-    core_model_to_hf_repo = {m.descriptor(): m.huggingface_repo for m in all_registered_models()}
+        "together": MODEL_ENTRIES,
-    default_models = [
+    }
-        ModelInput(
+    default_models = get_model_registry(available_models)
-            model_id=core_model_to_hf_repo[m.llama_model] if m.llama_model else m.provider_model_id,
-            provider_model_id=m.provider_model_id,
-            provider_id="together",
-            metadata=m.metadata,
-            model_type=m.model_type,
-        )
-        for m in MODEL_ENTRIES
-    ]
     default_tool_groups = [
         ToolGroupInput(
             toolgroup_id="builtin::websearch",
@@ -99,7 +90,7 @@ def get_distribution_template() -> DistributionTemplate:
         container_image=None,
         template_path=Path(__file__).parent / "doc_template.md",
         providers=providers,
-        default_models=default_models,
+        available_models_by_provider=available_models,
         run_configs={
             "run.yaml": RunConfigSettings(
                 provider_overrides={

llama_stack/templates/vllm-gpu/vllm.py

@@ -88,7 +88,6 @@ def get_distribution_template() -> DistributionTemplate:
         container_image=None,
         template_path=None,
         providers=providers,
-        default_models=[inference_model],
         run_configs={
             "run.yaml": RunConfigSettings(
                 provider_overrides={