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

cheesecake100201 · Nov 6, 2024 · 994732e · 994732e
1 parent b10e9f4
commit 994732e
Show file tree

Hide file tree

Showing 169 changed files with 106 additions and 105 deletions.
diff --git a/.gitmodules b/.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
diff --git a/docs/source/api_providers/new_api_provider.md b/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!

diff --git a/docs/source/getting_started/distributions/ondevice_distro/ios_sdk.md b/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

diff --git a/docs/source/getting_started/distributions/self_hosted_distro/ollama.md b/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`

diff --git a/docs/source/getting_started/index.md b/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`

diff --git a/llama_stack/providers/adapters/__init__.py → llama_stack/providers/inline/__init__.py b/llama_stack/providers/adapters/__init__.py → llama_stack/providers/inline/__init__.py
diff --git a/...ders/impls/braintrust/scoring/__init__.py → ...ers/inline/braintrust/scoring/__init__.py b/...ders/impls/braintrust/scoring/__init__.py → ...ers/inline/braintrust/scoring/__init__.py
diff --git a/...rs/impls/braintrust/scoring/braintrust.py → ...s/inline/braintrust/scoring/braintrust.py b/...rs/impls/braintrust/scoring/braintrust.py → ...s/inline/braintrust/scoring/braintrust.py
@@ -16,7 +16,7 @@
 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,
 )
 

diff --git a/...viders/impls/braintrust/scoring/config.py → ...iders/inline/braintrust/scoring/config.py b/...viders/impls/braintrust/scoring/config.py → ...iders/inline/braintrust/scoring/config.py
diff --git a/...ack/providers/adapters/agents/__init__.py → ...braintrust/scoring/scoring_fn/__init__.py b/...ack/providers/adapters/agents/__init__.py → ...braintrust/scoring/scoring_fn/__init__.py
diff --git a/.../providers/adapters/inference/__init__.py → ...st/scoring/scoring_fn/fn_defs/__init__.py b/.../providers/adapters/inference/__init__.py → ...st/scoring/scoring_fn/fn_defs/__init__.py
diff --git a/.../scoring_fn/fn_defs/answer_correctness.py → .../scoring_fn/fn_defs/answer_correctness.py b/.../scoring_fn/fn_defs/answer_correctness.py → .../scoring_fn/fn_defs/answer_correctness.py
diff --git a/.../scoring/scoring_fn/fn_defs/factuality.py → .../scoring/scoring_fn/fn_defs/factuality.py b/.../scoring/scoring_fn/fn_defs/factuality.py → .../scoring/scoring_fn/fn_defs/factuality.py
diff --git a/...alInferenceImpl.xcodeproj/project.pbxproj → ...alInferenceImpl.xcodeproj/project.pbxproj b/...alInferenceImpl.xcodeproj/project.pbxproj → ...alInferenceImpl.xcodeproj/project.pbxproj
diff --git a/...ject.xcworkspace/contents.xcworkspacedata → ...ject.xcworkspace/contents.xcworkspacedata b/...ject.xcworkspace/contents.xcworkspacedata → ...ject.xcworkspace/contents.xcworkspacedata
diff --git a/...ace/xcshareddata/IDEWorkspaceChecks.plist → ...ace/xcshareddata/IDEWorkspaceChecks.plist b/...ace/xcshareddata/IDEWorkspaceChecks.plist → ...ace/xcshareddata/IDEWorkspaceChecks.plist
diff --git a/...rence/LocalInferenceImpl/LocalInference.h → ...rence/LocalInferenceImpl/LocalInference.h b/...rence/LocalInferenceImpl/LocalInference.h → ...rence/LocalInferenceImpl/LocalInference.h
diff --git a/...e/LocalInferenceImpl/LocalInference.swift → ...e/LocalInferenceImpl/LocalInference.swift b/...e/LocalInferenceImpl/LocalInference.swift → ...e/LocalInferenceImpl/LocalInference.swift
diff --git a/...nference/LocalInferenceImpl/Parsing.swift → ...nference/LocalInferenceImpl/Parsing.swift b/...nference/LocalInferenceImpl/Parsing.swift → ...nference/LocalInferenceImpl/Parsing.swift
diff --git a/...e/LocalInferenceImpl/PromptTemplate.swift → ...e/LocalInferenceImpl/PromptTemplate.swift b/...e/LocalInferenceImpl/PromptTemplate.swift → ...e/LocalInferenceImpl/PromptTemplate.swift
diff --git a/...ce/LocalInferenceImpl/SystemPrompts.swift → ...ce/LocalInferenceImpl/SystemPrompts.swift b/...ce/LocalInferenceImpl/SystemPrompts.swift → ...ce/LocalInferenceImpl/SystemPrompts.swift
diff --git a/.../providers/impls/ios/inference/executorch → ...providers/inline/ios/inference/executorch b/.../providers/impls/ios/inference/executorch → ...providers/inline/ios/inference/executorch
diff --git a/...ack/providers/adapters/memory/__init__.py → ...oviders/inline/meta_reference/__init__.py b/...ack/providers/adapters/memory/__init__.py → ...oviders/inline/meta_reference/__init__.py
diff --git a/...s/impls/meta_reference/agents/__init__.py → .../inline/meta_reference/agents/__init__.py b/...s/impls/meta_reference/agents/__init__.py → .../inline/meta_reference/agents/__init__.py
diff --git a/...s/meta_reference/agents/agent_instance.py → ...e/meta_reference/agents/agent_instance.py b/...s/meta_reference/agents/agent_instance.py → ...e/meta_reference/agents/agent_instance.py
diff --git a/...ers/impls/meta_reference/agents/agents.py → ...rs/inline/meta_reference/agents/agents.py b/...ers/impls/meta_reference/agents/agents.py → ...rs/inline/meta_reference/agents/agents.py
diff --git a/...ers/impls/meta_reference/agents/config.py → ...rs/inline/meta_reference/agents/config.py b/...ers/impls/meta_reference/agents/config.py → ...rs/inline/meta_reference/agents/config.py
diff --git a/...mpls/meta_reference/agents/persistence.py → ...line/meta_reference/agents/persistence.py b/...mpls/meta_reference/agents/persistence.py → ...line/meta_reference/agents/persistence.py
diff --git a/...ack/providers/adapters/safety/__init__.py → ...ine/meta_reference/agents/rag/__init__.py b/...ack/providers/adapters/safety/__init__.py → ...ine/meta_reference/agents/rag/__init__.py
diff --git a/...reference/agents/rag/context_retriever.py → ...reference/agents/rag/context_retriever.py b/...reference/agents/rag/context_retriever.py → ...reference/agents/rag/context_retriever.py
diff --git a/...ers/impls/meta_reference/agents/safety.py → ...rs/inline/meta_reference/agents/safety.py b/...ers/impls/meta_reference/agents/safety.py → ...rs/inline/meta_reference/agents/safety.py
diff --git a/.../providers/adapters/telemetry/__init__.py → ...e/meta_reference/agents/tests/__init__.py b/.../providers/adapters/telemetry/__init__.py → ...e/meta_reference/agents/tests/__init__.py
diff --git a/..._reference/agents/tests/code_execution.py → ..._reference/agents/tests/code_execution.py b/..._reference/agents/tests/code_execution.py → ..._reference/agents/tests/code_execution.py
diff --git a/...reference/agents/tests/test_chat_agent.py → ...reference/agents/tests/test_chat_agent.py b/...reference/agents/tests/test_chat_agent.py → ...reference/agents/tests/test_chat_agent.py
diff --git a/llama_stack/providers/impls/__init__.py → ...e/meta_reference/agents/tools/__init__.py b/llama_stack/providers/impls/__init__.py → ...e/meta_reference/agents/tools/__init__.py
diff --git a/...impls/meta_reference/agents/tools/base.py → ...nline/meta_reference/agents/tools/base.py b/...impls/meta_reference/agents/tools/base.py → ...nline/meta_reference/agents/tools/base.py
diff --git a/...ls/meta_reference/agents/tools/builtin.py → ...ne/meta_reference/agents/tools/builtin.py b/...ls/meta_reference/agents/tools/builtin.py → ...ne/meta_reference/agents/tools/builtin.py
diff --git a/...braintrust/scoring/scoring_fn/__init__.py → ...nce/agents/tools/ipython_tool/__init__.py b/...braintrust/scoring/scoring_fn/__init__.py → ...nce/agents/tools/ipython_tool/__init__.py
diff --git a/...nts/tools/ipython_tool/code_env_prefix.py → ...nts/tools/ipython_tool/code_env_prefix.py b/...nts/tools/ipython_tool/code_env_prefix.py → ...nts/tools/ipython_tool/code_env_prefix.py
diff --git a/...ents/tools/ipython_tool/code_execution.py → ...ents/tools/ipython_tool/code_execution.py b/...ents/tools/ipython_tool/code_execution.py → ...ents/tools/ipython_tool/code_execution.py
diff --git a/...ipython_tool/matplotlib_custom_backend.py → ...ipython_tool/matplotlib_custom_backend.py b/...ipython_tool/matplotlib_custom_backend.py → ...ipython_tool/matplotlib_custom_backend.py
diff --git a/...erence/agents/tools/ipython_tool/utils.py → ...erence/agents/tools/ipython_tool/utils.py b/...erence/agents/tools/ipython_tool/utils.py → ...erence/agents/tools/ipython_tool/utils.py
diff --git a/...pls/meta_reference/agents/tools/safety.py → ...ine/meta_reference/agents/tools/safety.py b/...pls/meta_reference/agents/tools/safety.py → ...ine/meta_reference/agents/tools/safety.py
@@ -9,7 +9,7 @@
 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
 

diff --git a/...pls/meta_reference/codeshield/__init__.py → ...ine/meta_reference/codeshield/__init__.py b/...pls/meta_reference/codeshield/__init__.py → ...ine/meta_reference/codeshield/__init__.py
diff --git a/...meta_reference/codeshield/code_scanner.py → ...meta_reference/codeshield/code_scanner.py b/...meta_reference/codeshield/code_scanner.py → ...meta_reference/codeshield/code_scanner.py
diff --git a/...impls/meta_reference/codeshield/config.py → ...nline/meta_reference/codeshield/config.py b/...impls/meta_reference/codeshield/config.py → ...nline/meta_reference/codeshield/config.py
diff --git a/...mpls/meta_reference/datasetio/__init__.py → ...line/meta_reference/datasetio/__init__.py b/...mpls/meta_reference/datasetio/__init__.py → ...line/meta_reference/datasetio/__init__.py
diff --git a/.../impls/meta_reference/datasetio/config.py → ...inline/meta_reference/datasetio/config.py b/.../impls/meta_reference/datasetio/config.py → ...inline/meta_reference/datasetio/config.py
diff --git a/...pls/meta_reference/datasetio/datasetio.py → ...ine/meta_reference/datasetio/datasetio.py b/...pls/meta_reference/datasetio/datasetio.py → ...ine/meta_reference/datasetio/datasetio.py
diff --git a/...ers/impls/meta_reference/eval/__init__.py → ...rs/inline/meta_reference/eval/__init__.py b/...ers/impls/meta_reference/eval/__init__.py → ...rs/inline/meta_reference/eval/__init__.py
diff --git a/...iders/impls/meta_reference/eval/config.py → ...ders/inline/meta_reference/eval/config.py b/...iders/impls/meta_reference/eval/config.py → ...ders/inline/meta_reference/eval/config.py
diff --git a/...oviders/impls/meta_reference/eval/eval.py → ...viders/inline/meta_reference/eval/eval.py b/...oviders/impls/meta_reference/eval/eval.py → ...viders/inline/meta_reference/eval/eval.py
diff --git a/...mpls/meta_reference/inference/__init__.py → ...line/meta_reference/inference/__init__.py b/...mpls/meta_reference/inference/__init__.py → ...line/meta_reference/inference/__init__.py
diff --git a/.../impls/meta_reference/inference/config.py → ...inline/meta_reference/inference/config.py b/.../impls/meta_reference/inference/config.py → ...inline/meta_reference/inference/config.py
diff --git a/...ls/meta_reference/inference/generation.py → ...ne/meta_reference/inference/generation.py b/...ls/meta_reference/inference/generation.py → ...ne/meta_reference/inference/generation.py
diff --git a/...pls/meta_reference/inference/inference.py → ...ine/meta_reference/inference/inference.py b/...pls/meta_reference/inference/inference.py → ...ine/meta_reference/inference/inference.py
diff --git a/...eta_reference/inference/model_parallel.py → ...eta_reference/inference/model_parallel.py b/...eta_reference/inference/model_parallel.py → ...eta_reference/inference/model_parallel.py
diff --git a/...eta_reference/inference/parallel_utils.py → ...eta_reference/inference/parallel_utils.py b/...eta_reference/inference/parallel_utils.py → ...eta_reference/inference/parallel_utils.py
diff --git a/...st/scoring/scoring_fn/fn_defs/__init__.py → ...erence/inference/quantization/__init__.py b/...st/scoring/scoring_fn/fn_defs/__init__.py → ...erence/inference/quantization/__init__.py
diff --git a/...rence/inference/quantization/fp8_impls.py → ...rence/inference/quantization/fp8_impls.py b/...rence/inference/quantization/fp8_impls.py → ...rence/inference/quantization/fp8_impls.py
diff --git a/...erence/quantization/fp8_txest_disabled.py → ...erence/quantization/fp8_txest_disabled.py b/...erence/quantization/fp8_txest_disabled.py → ...erence/quantization/fp8_txest_disabled.py
diff --git a/.../inference/quantization/hadamard_utils.py → .../inference/quantization/hadamard_utils.py b/.../inference/quantization/hadamard_utils.py → .../inference/quantization/hadamard_utils.py
diff --git a/...eference/inference/quantization/loader.py → ...eference/inference/quantization/loader.py b/...eference/inference/quantization/loader.py → ...eference/inference/quantization/loader.py
@@ -27,7 +27,7 @@
 
 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,
 )
 

diff --git a/...roviders/impls/meta_reference/__init__.py → ...nference/quantization/scripts/__init__.py b/...roviders/impls/meta_reference/__init__.py → ...nference/quantization/scripts/__init__.py
diff --git a/...rence/quantization/scripts/build_conda.sh → ...rence/quantization/scripts/build_conda.sh b/...rence/quantization/scripts/build_conda.sh → ...rence/quantization/scripts/build_conda.sh
diff --git a/...antization/scripts/quantize_checkpoint.py → ...antization/scripts/quantize_checkpoint.py b/...antization/scripts/quantize_checkpoint.py → ...antization/scripts/quantize_checkpoint.py
diff --git a/...zation/scripts/run_quantize_checkpoint.sh → ...zation/scripts/run_quantize_checkpoint.sh b/...zation/scripts/run_quantize_checkpoint.sh → ...zation/scripts/run_quantize_checkpoint.sh
diff --git a/...s/impls/meta_reference/memory/__init__.py → .../inline/meta_reference/memory/__init__.py b/...s/impls/meta_reference/memory/__init__.py → .../inline/meta_reference/memory/__init__.py
diff --git a/...ers/impls/meta_reference/memory/config.py → ...rs/inline/meta_reference/memory/config.py b/...ers/impls/meta_reference/memory/config.py → ...rs/inline/meta_reference/memory/config.py
diff --git a/...ders/impls/meta_reference/memory/faiss.py → ...ers/inline/meta_reference/memory/faiss.py b/...ders/impls/meta_reference/memory/faiss.py → ...ers/inline/meta_reference/memory/faiss.py
diff --git a/...meta_reference/memory/tests/test_faiss.py → ...meta_reference/memory/tests/test_faiss.py b/...meta_reference/memory/tests/test_faiss.py → ...meta_reference/memory/tests/test_faiss.py
@@ -8,9 +8,9 @@
 
 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
 
 

diff --git a/...s/impls/meta_reference/safety/__init__.py → .../inline/meta_reference/safety/__init__.py b/...s/impls/meta_reference/safety/__init__.py → .../inline/meta_reference/safety/__init__.py
diff --git a/...iders/impls/meta_reference/safety/base.py → ...ders/inline/meta_reference/safety/base.py b/...iders/impls/meta_reference/safety/base.py → ...ders/inline/meta_reference/safety/base.py
diff --git a/...ers/impls/meta_reference/safety/config.py → ...rs/inline/meta_reference/safety/config.py b/...ers/impls/meta_reference/safety/config.py → ...rs/inline/meta_reference/safety/config.py
diff --git a/...mpls/meta_reference/safety/llama_guard.py → ...line/meta_reference/safety/llama_guard.py b/...mpls/meta_reference/safety/llama_guard.py → ...line/meta_reference/safety/llama_guard.py
diff --git a/...pls/meta_reference/safety/prompt_guard.py → ...ine/meta_reference/safety/prompt_guard.py b/...pls/meta_reference/safety/prompt_guard.py → ...ine/meta_reference/safety/prompt_guard.py
diff --git a/...ers/impls/meta_reference/safety/safety.py → ...rs/inline/meta_reference/safety/safety.py b/...ers/impls/meta_reference/safety/safety.py → ...rs/inline/meta_reference/safety/safety.py
diff --git a/.../impls/meta_reference/scoring/__init__.py → ...inline/meta_reference/scoring/__init__.py b/.../impls/meta_reference/scoring/__init__.py → ...inline/meta_reference/scoring/__init__.py
diff --git a/...rs/impls/meta_reference/scoring/config.py → ...s/inline/meta_reference/scoring/config.py b/...rs/impls/meta_reference/scoring/config.py → ...s/inline/meta_reference/scoring/config.py
diff --git a/...s/impls/meta_reference/scoring/scoring.py → .../inline/meta_reference/scoring/scoring.py b/...s/impls/meta_reference/scoring/scoring.py → .../inline/meta_reference/scoring/scoring.py
@@ -13,15 +13,15 @@
 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,
 )
 

diff --git a/...pls/meta_reference/agents/rag/__init__.py → ..._reference/scoring/scoring_fn/__init__.py b/...pls/meta_reference/agents/rag/__init__.py → ..._reference/scoring/scoring_fn/__init__.py
diff --git a/...nce/scoring/scoring_fn/base_scoring_fn.py → ...nce/scoring/scoring_fn/base_scoring_fn.py b/...nce/scoring/scoring_fn/base_scoring_fn.py → ...nce/scoring/scoring_fn/base_scoring_fn.py
diff --git a/...ta_reference/scoring/scoring_fn/common.py → ...ta_reference/scoring/scoring_fn/common.py b/...ta_reference/scoring/scoring_fn/common.py → ...ta_reference/scoring/scoring_fn/common.py
diff --git a/...scoring/scoring_fn/equality_scoring_fn.py → ...scoring/scoring_fn/equality_scoring_fn.py b/...scoring/scoring_fn/equality_scoring_fn.py → ...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,
 )
 

diff --git a/...s/meta_reference/agents/tests/__init__.py → ...ce/scoring/scoring_fn/fn_defs/__init__.py b/...s/meta_reference/agents/tests/__init__.py → ...ce/scoring/scoring_fn/fn_defs/__init__.py
diff --git a/...ce/scoring/scoring_fn/fn_defs/equality.py → ...ce/scoring/scoring_fn/fn_defs/equality.py b/...ce/scoring/scoring_fn/fn_defs/equality.py → ...ce/scoring/scoring_fn/fn_defs/equality.py
diff --git a/...fn/fn_defs/llm_as_judge_8b_correctness.py → ...fn/fn_defs/llm_as_judge_8b_correctness.py b/...fn/fn_defs/llm_as_judge_8b_correctness.py → ...fn/fn_defs/llm_as_judge_8b_correctness.py
diff --git a/...e/scoring/scoring_fn/fn_defs/subset_of.py → ...e/scoring/scoring_fn/fn_defs/subset_of.py b/...e/scoring/scoring_fn/fn_defs/subset_of.py → ...e/scoring/scoring_fn/fn_defs/subset_of.py
diff --git a/...ing/scoring_fn/llm_as_judge_scoring_fn.py → ...ing/scoring_fn/llm_as_judge_scoring_fn.py b/...ing/scoring_fn/llm_as_judge_scoring_fn.py → ...ing/scoring_fn/llm_as_judge_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.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
 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,
 )
 

diff --git a/...coring/scoring_fn/subset_of_scoring_fn.py → ...coring/scoring_fn/subset_of_scoring_fn.py b/...coring/scoring_fn/subset_of_scoring_fn.py → ...coring/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,
 )
 

diff --git a/...mpls/meta_reference/telemetry/__init__.py → ...line/meta_reference/telemetry/__init__.py b/...mpls/meta_reference/telemetry/__init__.py → ...line/meta_reference/telemetry/__init__.py
diff --git a/.../impls/meta_reference/telemetry/config.py → ...inline/meta_reference/telemetry/config.py b/.../impls/meta_reference/telemetry/config.py → ...inline/meta_reference/telemetry/config.py
diff --git a/...impls/meta_reference/telemetry/console.py → ...nline/meta_reference/telemetry/console.py b/...impls/meta_reference/telemetry/console.py → ...nline/meta_reference/telemetry/console.py
diff --git a/llama_stack/providers/impls/vllm/__init__.py → ...a_stack/providers/inline/vllm/__init__.py b/llama_stack/providers/impls/vllm/__init__.py → ...a_stack/providers/inline/vllm/__init__.py
diff --git a/llama_stack/providers/impls/vllm/config.py → llama_stack/providers/inline/vllm/config.py b/llama_stack/providers/impls/vllm/config.py → llama_stack/providers/inline/vllm/config.py
diff --git a/llama_stack/providers/impls/vllm/vllm.py → llama_stack/providers/inline/vllm/vllm.py b/llama_stack/providers/impls/vllm/vllm.py → llama_stack/providers/inline/vllm/vllm.py
diff --git a/llama_stack/providers/registry/agents.py b/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",
             ),
         ),
     ]
diff --git a/llama_stack/providers/registry/datasetio.py b/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=[],
         ),
     ]
diff --git a/llama_stack/providers/registry/eval.py b/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,

diff --git a/llama_stack/providers/registry/inference.py b/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,
@@ -40,61 +40,61 @@ 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(
             api=Api.inference,
             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(
             api=Api.inference,
             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(
             api=Api.inference,
             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(
             api=Api.inference,
             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(
             api=Api.inference,
             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(
@@ -115,18 +115,18 @@ 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(
             api=Api.inference,
             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",
         ),
     ]