impls -> inline, adapters -> remote (#381)

.gitmodules

@@ -1,3 +1,3 @@
 [submodule "llama_stack/providers/impls/ios/inference/executorch"]
-	path = llama_stack/providers/impls/ios/inference/executorch
+	path = llama_stack/providers/inline/ios/inference/executorch
 	url = https://github.com/pytorch/executorch

docs/source/api_providers/new_api_provider.md

@@ -6,8 +6,8 @@ This guide contains references to walk you through adding a new API provider.
 1. First, decide which API your provider falls into (e.g. Inference, Safety, Agents, Memory).
 2. Decide whether your provider is a remote provider, or inline implmentation. A remote provider is a provider that makes a remote request to an service. An inline provider is a provider where implementation is executed locally. Checkout the examples, and follow the structure to add your own API provider. Please find the following code pointers:
 
-    - [Inference Remote Adapter](https://github.com/meta-llama/llama-stack/tree/docs/llama_stack/providers/adapters/inference)
-    - [Inference Inline Provider](https://github.com/meta-llama/llama-stack/tree/docs/llama_stack/providers/impls/meta_reference/inference)
+    - [Inference Remote Adapter](https://github.com/meta-llama/llama-stack/tree/docs/llama_stack/providers/remote/inference)
+    - [Inference Inline Provider](https://github.com/meta-llama/llama-stack/tree/docs/llama_stack/providers/inline/meta_reference/inference)
 
 3. [Build a Llama Stack distribution](https://llama-stack.readthedocs.io/en/latest/distribution_dev/building_distro.html) with your API provider.
 4. Test your code!

docs/source/getting_started/distributions/ondevice_distro/ios_sdk.md

@@ -3,7 +3,7 @@
 We offer both remote and on-device use of Llama Stack in Swift via two components:
 
 1. [llama-stack-client-swift](https://github.com/meta-llama/llama-stack-client-swift/)
-2. [LocalInferenceImpl](https://github.com/meta-llama/llama-stack/tree/main/llama_stack/providers/impls/ios/inference)
+2. [LocalInferenceImpl](https://github.com/meta-llama/llama-stack/tree/main/llama_stack/providers/inline/ios/inference)
 
 ```{image} ../../../../_static/remote_or_local.gif
 :alt: Seamlessly switching between local, on-device inference and remote hosted inference

docs/source/getting_started/distributions/self_hosted_distro/ollama.md

@@ -102,7 +102,7 @@ ollama pull llama3.1:70b-instruct-fp16
 ```
 
 > [!NOTE]
-> Please check the [OLLAMA_SUPPORTED_MODELS](https://github.com/meta-llama/llama-stack/blob/main/llama_stack/providers/adapters/inference/ollama/ollama.py) for the supported Ollama models.
+> Please check the [OLLAMA_SUPPORTED_MODELS](https://github.com/meta-llama/llama-stack/blob/main/llama_stack/providers.remote/inference/ollama/ollama.py) for the supported Ollama models.
 
 
 To serve a new model with `ollama`

docs/source/getting_started/index.md

@@ -386,7 +386,7 @@ ollama pull llama3.1:8b-instruct-fp16
 ollama pull llama3.1:70b-instruct-fp16
 ```
 
-> Please check the [OLLAMA_SUPPORTED_MODELS](https://github.com/meta-llama/llama-stack/blob/main/llama_stack/providers/adapters/inference/ollama/ollama.py) for the supported Ollama models.
+> Please check the [OLLAMA_SUPPORTED_MODELS](https://github.com/meta-llama/llama-stack/blob/main/llama_stack/providers.remote/inference/ollama/ollama.py) for the supported Ollama models.
 
 
 To serve a new model with `ollama`

llama_stack/providers/inline/braintrust/scoring/braintrust.py

@@ -16,7 +16,7 @@ from llama_stack.apis.datasets import *  # noqa: F403
 from autoevals.llm import Factuality
 from autoevals.ragas import AnswerCorrectness
 from llama_stack.providers.datatypes import ScoringFunctionsProtocolPrivate
-from llama_stack.providers.impls.meta_reference.scoring.scoring_fn.common import (
+from llama_stack.providers.inline.meta_reference.scoring.scoring_fn.common import (
     aggregate_average,
 )
 

llama_stack/providers/inline/meta_reference/agents/tools/safety.py

@@ -9,7 +9,7 @@ from typing import List
 from llama_stack.apis.inference import Message
 from llama_stack.apis.safety import *  # noqa: F403
 
-from llama_stack.providers.impls.meta_reference.agents.safety import ShieldRunnerMixin
+from llama_stack.providers.inline.meta_reference.agents.safety import ShieldRunnerMixin
 
 from .builtin import BaseTool
 

llama_stack/providers/inline/meta_reference/inference/quantization/loader.py

@@ -27,7 +27,7 @@ from torchao.quantization.GPTQ import Int8DynActInt4WeightLinear
 
 from llama_stack.apis.inference import QuantizationType
 
-from llama_stack.providers.impls.meta_reference.inference.config import (
+from llama_stack.providers.inline.meta_reference.inference.config import (
     MetaReferenceQuantizedInferenceConfig,
 )
 

llama_stack/providers/inline/meta_reference/memory/tests/test_faiss.py

@@ -8,9 +8,9 @@ import tempfile
 
 import pytest
 from llama_stack.apis.memory import MemoryBankType, VectorMemoryBankDef
-from llama_stack.providers.impls.meta_reference.memory.config import FaissImplConfig
+from llama_stack.providers.inline.meta_reference.memory.config import FaissImplConfig
 
-from llama_stack.providers.impls.meta_reference.memory.faiss import FaissMemoryImpl
+from llama_stack.providers.inline.meta_reference.memory.faiss import FaissMemoryImpl
 from llama_stack.providers.utils.kvstore.config import SqliteKVStoreConfig
 
 

llama_stack/providers/inline/meta_reference/scoring/scoring.py

@@ -13,15 +13,15 @@ from llama_stack.apis.datasetio import *  # noqa: F403
 from llama_stack.apis.datasets import *  # noqa: F403
 from llama_stack.apis.inference.inference import Inference
 from llama_stack.providers.datatypes import ScoringFunctionsProtocolPrivate
-from llama_stack.providers.impls.meta_reference.scoring.scoring_fn.equality_scoring_fn import (
+from llama_stack.providers.inline.meta_reference.scoring.scoring_fn.equality_scoring_fn import (
     EqualityScoringFn,
 )
 
-from llama_stack.providers.impls.meta_reference.scoring.scoring_fn.llm_as_judge_scoring_fn import (
+from llama_stack.providers.inline.meta_reference.scoring.scoring_fn.llm_as_judge_scoring_fn import (
     LlmAsJudgeScoringFn,
 )
 
-from llama_stack.providers.impls.meta_reference.scoring.scoring_fn.subset_of_scoring_fn import (
+from llama_stack.providers.inline.meta_reference.scoring.scoring_fn.subset_of_scoring_fn import (
     SubsetOfScoringFn,
 )
 

llama_stack/providers/inline/meta_reference/scoring/scoring_fn/equality_scoring_fn.py

@@ -4,18 +4,18 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack.providers.impls.meta_reference.scoring.scoring_fn.base_scoring_fn import (
+from llama_stack.providers.inline.meta_reference.scoring.scoring_fn.base_scoring_fn import (
     BaseScoringFn,
 )
 from llama_stack.apis.scoring_functions import *  # noqa: F401, F403
 from llama_stack.apis.scoring import *  # noqa: F401, F403
 from llama_stack.apis.common.type_system import *  # noqa: F403
 
-from llama_stack.providers.impls.meta_reference.scoring.scoring_fn.common import (
+from llama_stack.providers.inline.meta_reference.scoring.scoring_fn.common import (
     aggregate_accuracy,
 )
 
-from llama_stack.providers.impls.meta_reference.scoring.scoring_fn.fn_defs.equality import (
+from llama_stack.providers.inline.meta_reference.scoring.scoring_fn.fn_defs.equality import (
     equality,
 )
 

llama_stack/providers/inline/meta_reference/scoring/scoring_fn/llm_as_judge_scoring_fn.py

@@ -4,7 +4,7 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 from llama_stack.apis.inference.inference import Inference
-from llama_stack.providers.impls.meta_reference.scoring.scoring_fn.base_scoring_fn import (
+from llama_stack.providers.inline.meta_reference.scoring.scoring_fn.base_scoring_fn import (
     BaseScoringFn,
 )
 from llama_stack.apis.scoring_functions import *  # noqa: F401, F403
@@ -12,10 +12,10 @@ from llama_stack.apis.scoring import *  # noqa: F401, F403
 from llama_stack.apis.common.type_system import *  # noqa: F403
 import re
 
-from llama_stack.providers.impls.meta_reference.scoring.scoring_fn.common import (
+from llama_stack.providers.inline.meta_reference.scoring.scoring_fn.common import (
     aggregate_average,
 )
-from llama_stack.providers.impls.meta_reference.scoring.scoring_fn.fn_defs.llm_as_judge_8b_correctness import (
+from llama_stack.providers.inline.meta_reference.scoring.scoring_fn.fn_defs.llm_as_judge_8b_correctness import (
     llm_as_judge_8b_correctness,
 )
 

llama_stack/providers/inline/meta_reference/scoring/scoring_fn/subset_of_scoring_fn.py

@@ -4,17 +4,17 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack.providers.impls.meta_reference.scoring.scoring_fn.base_scoring_fn import (
+from llama_stack.providers.inline.meta_reference.scoring.scoring_fn.base_scoring_fn import (
     BaseScoringFn,
 )
 from llama_stack.apis.scoring_functions import *  # noqa: F401, F403
 from llama_stack.apis.scoring import *  # noqa: F401, F403
 from llama_stack.apis.common.type_system import *  # noqa: F403
-from llama_stack.providers.impls.meta_reference.scoring.scoring_fn.common import (
+from llama_stack.providers.inline.meta_reference.scoring.scoring_fn.common import (
     aggregate_accuracy,
 )
 
-from llama_stack.providers.impls.meta_reference.scoring.scoring_fn.fn_defs.subset_of import (
+from llama_stack.providers.inline.meta_reference.scoring.scoring_fn.fn_defs.subset_of import (
     subset_of,
 )
 

llama_stack/providers/registry/agents.py

@@ -22,8 +22,8 @@ def available_providers() -> List[ProviderSpec]:
                 "scikit-learn",
             ]
             + kvstore_dependencies(),
-            module="llama_stack.providers.impls.meta_reference.agents",
-            config_class="llama_stack.providers.impls.meta_reference.agents.MetaReferenceAgentsImplConfig",
+            module="llama_stack.providers.inline.meta_reference.agents",
+            config_class="llama_stack.providers.inline.meta_reference.agents.MetaReferenceAgentsImplConfig",
             api_dependencies=[
                 Api.inference,
                 Api.safety,
@@ -36,8 +36,8 @@ def available_providers() -> List[ProviderSpec]:
             adapter=AdapterSpec(
                 adapter_type="sample",
                 pip_packages=[],
-                module="llama_stack.providers.adapters.agents.sample",
-                config_class="llama_stack.providers.adapters.agents.sample.SampleConfig",
+                module="llama_stack.providers.remote.agents.sample",
+                config_class="llama_stack.providers.remote.agents.sample.SampleConfig",
             ),
         ),
     ]

llama_stack/providers/registry/datasetio.py

@@ -15,8 +15,8 @@ def available_providers() -> List[ProviderSpec]:
             api=Api.datasetio,
             provider_type="meta-reference",
             pip_packages=["pandas"],
-            module="llama_stack.providers.impls.meta_reference.datasetio",
-            config_class="llama_stack.providers.impls.meta_reference.datasetio.MetaReferenceDatasetIOConfig",
+            module="llama_stack.providers.inline.meta_reference.datasetio",
+            config_class="llama_stack.providers.inline.meta_reference.datasetio.MetaReferenceDatasetIOConfig",
             api_dependencies=[],
         ),
     ]

llama_stack/providers/registry/eval.py

@@ -15,8 +15,8 @@ def available_providers() -> List[ProviderSpec]:
             api=Api.eval,
             provider_type="meta-reference",
             pip_packages=[],
-            module="llama_stack.providers.impls.meta_reference.eval",
-            config_class="llama_stack.providers.impls.meta_reference.eval.MetaReferenceEvalConfig",
+            module="llama_stack.providers.inline.meta_reference.eval",
+            config_class="llama_stack.providers.inline.meta_reference.eval.MetaReferenceEvalConfig",
             api_dependencies=[
                 Api.datasetio,
                 Api.datasets,

llama_stack/providers/registry/inference.py

@@ -27,8 +27,8 @@ def available_providers() -> List[ProviderSpec]:
             api=Api.inference,
             provider_type="meta-reference",
             pip_packages=META_REFERENCE_DEPS,
-            module="llama_stack.providers.impls.meta_reference.inference",
-            config_class="llama_stack.providers.impls.meta_reference.inference.MetaReferenceInferenceConfig",
+            module="llama_stack.providers.inline.meta_reference.inference",
+            config_class="llama_stack.providers.inline.meta_reference.inference.MetaReferenceInferenceConfig",
         ),
         InlineProviderSpec(
             api=Api.inference,
@@ -40,16 +40,16 @@ def available_providers() -> List[ProviderSpec]:
                     "torchao==0.5.0",
                 ]
             ),
-            module="llama_stack.providers.impls.meta_reference.inference",
-            config_class="llama_stack.providers.impls.meta_reference.inference.MetaReferenceQuantizedInferenceConfig",
+            module="llama_stack.providers.inline.meta_reference.inference",
+            config_class="llama_stack.providers.inline.meta_reference.inference.MetaReferenceQuantizedInferenceConfig",
         ),
         remote_provider_spec(
             api=Api.inference,
             adapter=AdapterSpec(
                 adapter_type="sample",
                 pip_packages=[],
-                module="llama_stack.providers.adapters.inference.sample",
-                config_class="llama_stack.providers.adapters.inference.sample.SampleConfig",
+                module="llama_stack.providers.remote.inference.sample",
+                config_class="llama_stack.providers.remote.inference.sample.SampleConfig",
             ),
         ),
         remote_provider_spec(
@@ -57,8 +57,8 @@ def available_providers() -> List[ProviderSpec]:
             adapter=AdapterSpec(
                 adapter_type="ollama",
                 pip_packages=["ollama", "aiohttp"],
-                config_class="llama_stack.providers.adapters.inference.ollama.OllamaImplConfig",
-                module="llama_stack.providers.adapters.inference.ollama",
+                config_class="llama_stack.providers.remote.inference.ollama.OllamaImplConfig",
+                module="llama_stack.providers.remote.inference.ollama",
             ),
         ),
         remote_provider_spec(
@@ -66,8 +66,8 @@ def available_providers() -> List[ProviderSpec]:
             adapter=AdapterSpec(
                 adapter_type="vllm",
                 pip_packages=["openai"],
-                module="llama_stack.providers.adapters.inference.vllm",
-                config_class="llama_stack.providers.adapters.inference.vllm.VLLMInferenceAdapterConfig",
+                module="llama_stack.providers.remote.inference.vllm",
+                config_class="llama_stack.providers.remote.inference.vllm.VLLMInferenceAdapterConfig",
             ),
         ),
         remote_provider_spec(
@@ -75,8 +75,8 @@ def available_providers() -> List[ProviderSpec]:
             adapter=AdapterSpec(
                 adapter_type="tgi",
                 pip_packages=["huggingface_hub", "aiohttp"],
-                module="llama_stack.providers.adapters.inference.tgi",
-                config_class="llama_stack.providers.adapters.inference.tgi.TGIImplConfig",
+                module="llama_stack.providers.remote.inference.tgi",
+                config_class="llama_stack.providers.remote.inference.tgi.TGIImplConfig",
             ),
         ),
         remote_provider_spec(
@@ -84,8 +84,8 @@ def available_providers() -> List[ProviderSpec]:
             adapter=AdapterSpec(
                 adapter_type="hf::serverless",
                 pip_packages=["huggingface_hub", "aiohttp"],
-                module="llama_stack.providers.adapters.inference.tgi",
-                config_class="llama_stack.providers.adapters.inference.tgi.InferenceAPIImplConfig",
+                module="llama_stack.providers.remote.inference.tgi",
+                config_class="llama_stack.providers.remote.inference.tgi.InferenceAPIImplConfig",
             ),
         ),
         remote_provider_spec(
@@ -93,8 +93,8 @@ def available_providers() -> List[ProviderSpec]:
             adapter=AdapterSpec(
                 adapter_type="hf::endpoint",
                 pip_packages=["huggingface_hub", "aiohttp"],
-                module="llama_stack.providers.adapters.inference.tgi",
-                config_class="llama_stack.providers.adapters.inference.tgi.InferenceEndpointImplConfig",
+                module="llama_stack.providers.remote.inference.tgi",
+                config_class="llama_stack.providers.remote.inference.tgi.InferenceEndpointImplConfig",
             ),
         ),
         remote_provider_spec(
@@ -104,8 +104,8 @@ def available_providers() -> List[ProviderSpec]:
                 pip_packages=[
                     "fireworks-ai",
                 ],
-                module="llama_stack.providers.adapters.inference.fireworks",
-                config_class="llama_stack.providers.adapters.inference.fireworks.FireworksImplConfig",
+                module="llama_stack.providers.remote.inference.fireworks",
+                config_class="llama_stack.providers.remote.inference.fireworks.FireworksImplConfig",
             ),
         ),
         remote_provider_spec(
@@ -115,9 +115,9 @@ def available_providers() -> List[ProviderSpec]:
                 pip_packages=[
                     "together",
                 ],
-                module="llama_stack.providers.adapters.inference.together",
-                config_class="llama_stack.providers.adapters.inference.together.TogetherImplConfig",
-                provider_data_validator="llama_stack.providers.adapters.safety.together.TogetherProviderDataValidator",
+                module="llama_stack.providers.remote.inference.together",
+                config_class="llama_stack.providers.remote.inference.together.TogetherImplConfig",
+                provider_data_validator="llama_stack.providers.remote.safety.together.TogetherProviderDataValidator",
             ),
         ),
         remote_provider_spec(
@@ -125,8 +125,8 @@ def available_providers() -> List[ProviderSpec]:
             adapter=AdapterSpec(
                 adapter_type="bedrock",
                 pip_packages=["boto3"],
-                module="llama_stack.providers.adapters.inference.bedrock",
-                config_class="llama_stack.providers.adapters.inference.bedrock.BedrockConfig",
+                module="llama_stack.providers.remote.inference.bedrock",
+                config_class="llama_stack.providers.remote.inference.bedrock.BedrockConfig",
             ),
         ),
         remote_provider_spec(
@@ -136,8 +136,8 @@ def available_providers() -> List[ProviderSpec]:
                 pip_packages=[
                     "openai",
                 ],
-                module="llama_stack.providers.adapters.inference.databricks",
-                config_class="llama_stack.providers.adapters.inference.databricks.DatabricksImplConfig",
+                module="llama_stack.providers.remote.inference.databricks",
+                config_class="llama_stack.providers.remote.inference.databricks.DatabricksImplConfig",
             ),
         ),
         InlineProviderSpec(
@@ -146,7 +146,7 @@ def available_providers() -> List[ProviderSpec]:
             pip_packages=[
                 "vllm",
             ],
-            module="llama_stack.providers.impls.vllm",
-            config_class="llama_stack.providers.impls.vllm.VLLMConfig",
+            module="llama_stack.providers.inline.vllm",
+            config_class="llama_stack.providers.inline.vllm.VLLMConfig",
         ),
     ]