diff --git a/sdks/python/README.md b/sdks/python/README.md
index 9a1e80c..faf479a 100644
--- a/sdks/python/README.md
+++ b/sdks/python/README.md
@@ -8,7 +8,7 @@ Python SDK for Learning Commons educational text evaluators.
 - **`schemas/`** – Pydantic schemas for inputs, outputs, config, metadata, and errors.
 - **`providers/`** – LangChain-based LLM provider factory (OpenAI, Google, Anthropic).
 - **`settings/`** – TOML settings loader for evaluator configuration.
-- **`config.py`** – Re-exports config types (`EvaluatorConfig`, `PromptProviderConfig`, factory functions).
+- **`config.py`** – Re-exports config types (`EvaluatorConfig`, `LLMProviderConfig`, factory functions).
 - **`errors.py`** – Re-exports error types.
 - **`logger.py`** – Standard Python logging utilities.
 
@@ -138,7 +138,7 @@ import logging
 from learning_commons_evaluators import (
     ConventionalityEvaluator,
     ConventionalityEvaluationInput,
-    GooglePromptProviderConfig,
+    GoogleLLMProviderConfig,
     create_config,
 )
 
@@ -147,7 +147,7 @@ logging.basicConfig(level=logging.INFO)
 
 # Create config with provider credentials
 config = create_config(
-    google_prompt_provider_config=GooglePromptProviderConfig(api_key="your-google-key"),
+    google_llm_provider_config=GoogleLLMProviderConfig(api_key="your-google-key"),
     telemetry_partner_id="your-telemetry-id",
 )
 
@@ -175,12 +175,12 @@ Evaluates text for conventionality of language (idioms, metaphors, implied meani
 from learning_commons_evaluators import (
     ConventionalityEvaluator,
     ConventionalityEvaluationInput,
-    GooglePromptProviderConfig,
+    GoogleLLMProviderConfig,
     create_config,
 )
 
 config = create_config(
-    google_prompt_provider_config=GooglePromptProviderConfig(api_key="..."),
+    google_llm_provider_config=GoogleLLMProviderConfig(api_key="..."),
     telemetry_partner_id="your-telemetry-id",
 )
 evaluator = ConventionalityEvaluator(config)
@@ -205,14 +205,14 @@ Each LLM provider requires its own config with an API key:
 
 ```python
 from learning_commons_evaluators import (
-    GooglePromptProviderConfig,
-    OpenAIPromptProviderConfig,
-    AnthropicPromptProviderConfig,
+    GoogleLLMProviderConfig,
+    OpenAILLMProviderConfig,
+    AnthropicLLMProviderConfig,
 )
 
-google_config = GooglePromptProviderConfig(api_key="...")
-openai_config = OpenAIPromptProviderConfig(api_key="...")
-anthropic_config = AnthropicPromptProviderConfig(api_key="...")
+google_config = GoogleLLMProviderConfig(api_key="...")
+openai_config = OpenAILLMProviderConfig(api_key="...")
+anthropic_config = AnthropicLLMProviderConfig(api_key="...")
 ```
 
 ### EvaluatorConfig
@@ -223,8 +223,8 @@ Use `create_config()` to create an `EvaluatorConfig`:
 from learning_commons_evaluators import create_config
 
 config = create_config(
-    google_prompt_provider_config=google_config,
-    openai_prompt_provider_config=openai_config,
+    google_llm_provider_config=google_config,
+    openai_llm_provider_config=openai_config,
     telemetry_partner_id="your-telemetry-id",
     logger=my_logger,  # Optional: any standard logging.Logger (default: package logger)
 )
diff --git a/sdks/python/src/learning_commons_evaluators/__init__.py b/sdks/python/src/learning_commons_evaluators/__init__.py
index b3208dc..cccf241 100644
--- a/sdks/python/src/learning_commons_evaluators/__init__.py
+++ b/sdks/python/src/learning_commons_evaluators/__init__.py
@@ -4,11 +4,11 @@
 
 # Config
 from learning_commons_evaluators.config import (
-    AnthropicPromptProviderConfig,
+    AnthropicLLMProviderConfig,
     EvaluatorConfig,
-    GooglePromptProviderConfig,
-    OpenAIPromptProviderConfig,
-    PromptProviderConfig,
+    GoogleLLMProviderConfig,
+    LLMProviderConfig,
+    OpenAILLMProviderConfig,
     PromptSettings,
     TelemetryConfig,
     create_config,
@@ -55,7 +55,7 @@
     GradeInputField,
     TextInputField,
 )
-from learning_commons_evaluators.schemas.config import EvaluationSettings, LlmProvider
+from learning_commons_evaluators.schemas.config import EvaluationSettings, LLMProvider
 from learning_commons_evaluators.schemas.conventionality import (
     ConventionalityEvaluationSettings,
     ConventionalityOutput,
@@ -95,7 +95,7 @@
     "__description__",
     "__version__",
     "APIError",
-    "AnthropicPromptProviderConfig",
+    "AnthropicLLMProviderConfig",
     "AuthenticationError",
     "BaseEvaluator",
     "ConfigurationError",
@@ -115,7 +115,7 @@
     "EvaluatorMetadata",
     "EvaluatorRetryableError",
     "EvaluatorTimeoutError",
-    "GooglePromptProviderConfig",
+    "GoogleLLMProviderConfig",
     "AnyInputSpec",
     "GradeInputField",
     "GradeInputSpec",
@@ -123,12 +123,12 @@
     "InputSpec",
     "InputT",
     "TextInputSpec",
-    "LlmProvider",
+    "LLMProvider",
+    "LLMProviderConfig",
     "Logger",
     "NetworkError",
-    "OpenAIPromptProviderConfig",
+    "OpenAILLMProviderConfig",
     "OutputT",
-    "PromptProviderConfig",
     "PromptSettings",
     "RateLimitError",
     "SDK_LOGGER_NAME",
diff --git a/sdks/python/src/learning_commons_evaluators/config.py b/sdks/python/src/learning_commons_evaluators/config.py
index dc5995f..66e5d88 100644
--- a/sdks/python/src/learning_commons_evaluators/config.py
+++ b/sdks/python/src/learning_commons_evaluators/config.py
@@ -1,11 +1,11 @@
 """Re-export evaluator config from schemas.config for package-level imports."""
 
 from learning_commons_evaluators.schemas.config import (
-    AnthropicPromptProviderConfig,
+    AnthropicLLMProviderConfig,
     EvaluatorConfig,
-    GooglePromptProviderConfig,
-    OpenAIPromptProviderConfig,
-    PromptProviderConfig,
+    GoogleLLMProviderConfig,
+    LLMProviderConfig,
+    OpenAILLMProviderConfig,
     PromptSettings,
     TelemetryConfig,
     create_config,
@@ -14,11 +14,11 @@
 )
 
 __all__ = [
-    "AnthropicPromptProviderConfig",
+    "AnthropicLLMProviderConfig",
     "EvaluatorConfig",
-    "GooglePromptProviderConfig",
-    "OpenAIPromptProviderConfig",
-    "PromptProviderConfig",
+    "GoogleLLMProviderConfig",
+    "OpenAILLMProviderConfig",
+    "LLMProviderConfig",
     "PromptSettings",
     "TelemetryConfig",
     "create_config",
diff --git a/sdks/python/src/learning_commons_evaluators/evaluators/__init__.py b/sdks/python/src/learning_commons_evaluators/evaluators/__init__.py
index 5bb9fc1..83211f7 100644
--- a/sdks/python/src/learning_commons_evaluators/evaluators/__init__.py
+++ b/sdks/python/src/learning_commons_evaluators/evaluators/__init__.py
@@ -1,6 +1,6 @@
 """Evaluator implementations.
 
-Config types (``EvaluatorConfig``, ``GooglePromptProviderConfig``, etc.) are
+Config types (``EvaluatorConfig``, ``GoogleLLMProviderConfig``, etc.) are
 exported from the top-level :mod:`learning_commons_evaluators` package, not
 from here.  Import evaluator classes directly from this sub-package only when
 you want to be explicit about the source.
diff --git a/sdks/python/src/learning_commons_evaluators/providers/__init__.py b/sdks/python/src/learning_commons_evaluators/providers/__init__.py
index f30b238..fc42968 100644
--- a/sdks/python/src/learning_commons_evaluators/providers/__init__.py
+++ b/sdks/python/src/learning_commons_evaluators/providers/__init__.py
@@ -2,8 +2,8 @@
 
 create_provider() returns a langchain_core BaseChatModel (OpenAI, Google, Anthropic).
 
-Config types (PromptProviderConfig and provider-specific configs:
-AnthropicPromptProviderConfig, GooglePromptProviderConfig, OpenAIPromptProviderConfig)
+Config types (LLMProviderConfig and provider-specific configs:
+AnthropicLLMProviderConfig, GoogleLLMProviderConfig, OpenAILLMProviderConfig)
 live in learning_commons_evaluators.schemas.config.
 """
 
diff --git a/sdks/python/src/learning_commons_evaluators/providers/langchain.py b/sdks/python/src/learning_commons_evaluators/providers/langchain.py
index 9496831..27a9cc8 100644
--- a/sdks/python/src/learning_commons_evaluators/providers/langchain.py
+++ b/sdks/python/src/learning_commons_evaluators/providers/langchain.py
@@ -14,7 +14,7 @@
 from learning_commons_evaluators.errors import ConfigurationError
 from learning_commons_evaluators.schemas.config import (
     EvaluatorConfig,
-    LlmProvider,
+    LLMProvider,
     PromptSettings,
 )
 from learning_commons_evaluators.schemas.metadata import TokenUsage
@@ -48,7 +48,7 @@ def _create_openai_llm(
 ) -> BaseChatModel:
     from langchain_openai import ChatOpenAI
 
-    openai_config = evaluator_config.openai_prompt_provider_config
+    openai_config = evaluator_config.openai_llm_provider_config
     if openai_config is None:
         raise ConfigurationError("OpenAI provider config is not set on EvaluatorConfig")
     kwargs: dict[str, Any] = {
@@ -64,7 +64,7 @@ def _create_google_llm(
 ) -> BaseChatModel:
     from langchain_google_genai import ChatGoogleGenerativeAI
 
-    google_config = evaluator_config.google_prompt_provider_config
+    google_config = evaluator_config.google_llm_provider_config
     if google_config is None:
         raise ConfigurationError("Google provider config is not set on EvaluatorConfig")
     return ChatGoogleGenerativeAI(
@@ -79,7 +79,7 @@ def _create_anthropic_llm(
 ) -> BaseChatModel:
     from langchain_anthropic import ChatAnthropic
 
-    anthropic_config = evaluator_config.anthropic_prompt_provider_config
+    anthropic_config = evaluator_config.anthropic_llm_provider_config
     if anthropic_config is None:
         raise ConfigurationError("Anthropic provider config is not set on EvaluatorConfig")
     return ChatAnthropic(
@@ -93,11 +93,11 @@ def create_provider(
     prompt_settings: PromptSettings, evaluator_config: EvaluatorConfig
 ) -> BaseChatModel:
     """Create a LangChain chat model from a PromptSettings for use in a chain."""
-    if prompt_settings.provider_type == LlmProvider.OPENAI:
+    if prompt_settings.provider_type == LLMProvider.OPENAI:
         return _create_openai_llm(prompt_settings, evaluator_config)
-    if prompt_settings.provider_type == LlmProvider.GOOGLE:
+    if prompt_settings.provider_type == LLMProvider.GOOGLE:
         return _create_google_llm(prompt_settings, evaluator_config)
-    if prompt_settings.provider_type == LlmProvider.ANTHROPIC:
+    if prompt_settings.provider_type == LLMProvider.ANTHROPIC:
         return _create_anthropic_llm(prompt_settings, evaluator_config)
     raise ConfigurationError(
         f"Unsupported LLM provider type: {prompt_settings.provider_type!r}. "
diff --git a/sdks/python/src/learning_commons_evaluators/schemas/__init__.py b/sdks/python/src/learning_commons_evaluators/schemas/__init__.py
index 5070c69..b168e59 100644
--- a/sdks/python/src/learning_commons_evaluators/schemas/__init__.py
+++ b/sdks/python/src/learning_commons_evaluators/schemas/__init__.py
@@ -6,7 +6,7 @@
 )
 from learning_commons_evaluators.schemas.config import (
     EvaluationSettings,
-    LlmProvider,
+    LLMProvider,
     PromptSettings,
 )
 from learning_commons_evaluators.schemas.conventionality import (
@@ -61,7 +61,7 @@
     "GradeInputField",
     "InputField",
     "InputMetadata",
-    "LlmProvider",
+    "LLMProvider",
     "PromptSettings",
     "PROMPT_STEP_EXTRA_PROMPT_SETTINGS",
     "PROMPT_STEP_EXTRA_TOKEN_USAGE",
diff --git a/sdks/python/src/learning_commons_evaluators/schemas/config.py b/sdks/python/src/learning_commons_evaluators/schemas/config.py
index 94e44f2..8d1d035 100644
--- a/sdks/python/src/learning_commons_evaluators/schemas/config.py
+++ b/sdks/python/src/learning_commons_evaluators/schemas/config.py
@@ -1,7 +1,7 @@
 """
 Evaluator configuration and metadata.
 
-Prompt provider configs (base, Google, OpenAI, Anthropic), EvaluatorConfig, and evaluator
+LLM provider configs (base, Google, OpenAI, Anthropic), EvaluatorConfig, and evaluator
 metadata live here. Evaluator config is created via factory methods (create_config,
 create_config_no_telemetry, create_config_telemetry_with_full_input).
 """
@@ -13,11 +13,10 @@
 
 from learning_commons_evaluators.logger import Logger, get_logger
 
-# --- Prompt provider configs (for LLM calls in prompt steps) ---
+# --- LLM provider configs (for LLM calls in prompt steps) ---
 
 
-# TODO: rename to LLMProvider
-class LlmProvider(str, Enum):
+class LLMProvider(str, Enum):
     """LLM provider identifier. Subclass of str so it compares and serializes as the provider name."""
 
     ANTHROPIC = "anthropic"
@@ -25,27 +24,26 @@ class LlmProvider(str, Enum):
     OPENAI = "openai"
 
 
-# TODO: rename to LLMProviderConfig and subclasses to GoogleLLMProviderConfig, OpenAILLMProviderConfig, AnthropicLLMProviderConfig.
 @dataclass(frozen=True)
-class PromptProviderConfig:
-    """Base type for prompt provider configuration."""
+class LLMProviderConfig:
+    """Base type for LLM provider configuration."""
 
     api_key: str
-    type: LlmProvider
+    type: LLMProvider
 
 
 @dataclass(frozen=True)
-class GooglePromptProviderConfig(PromptProviderConfig):
-    """Google (Gemini) prompt provider config. Takes an API key."""
+class GoogleLLMProviderConfig(LLMProviderConfig):
+    """Google (Gemini) LLM provider config. Takes an API key."""
 
-    type: LlmProvider = LlmProvider.GOOGLE
+    type: LLMProvider = LLMProvider.GOOGLE
 
 
 @dataclass(frozen=True)
-class OpenAIPromptProviderConfig(PromptProviderConfig):
-    """OpenAI prompt provider config. Takes an API key."""
+class OpenAILLMProviderConfig(LLMProviderConfig):
+    """OpenAI LLM provider config. Takes an API key."""
 
-    type: LlmProvider = LlmProvider.OPENAI
+    type: LLMProvider = LLMProvider.OPENAI
     # TODO: verify base_url functionality before enabling
     # base_url: str | None = (
     #     None  # Optional; for OpenAI-compatible endpoints (e.g. Azure, proxy). Used only when type is OPENAI.
@@ -53,17 +51,17 @@ class OpenAIPromptProviderConfig(PromptProviderConfig):
 
 
 @dataclass(frozen=True)
-class AnthropicPromptProviderConfig(PromptProviderConfig):
-    """Anthropic (Claude) prompt provider config. Takes an API key."""
+class AnthropicLLMProviderConfig(LLMProviderConfig):
+    """Anthropic (Claude) LLM provider config. Takes an API key."""
 
-    type: LlmProvider = LlmProvider.ANTHROPIC
+    type: LLMProvider = LLMProvider.ANTHROPIC
 
 
 @dataclass(frozen=True)
 class PromptSettings:
     """Settings for a prompt step: provider, model, temperature."""
 
-    provider_type: LlmProvider
+    provider_type: LLMProvider
     model: str
     temperature: float
 
@@ -91,7 +89,7 @@ class TelemetryConfig:
 @dataclass(frozen=True)
 class EvaluatorConfig:
     """
-    Config for creating an evaluator: prompt provider configs, logger, telemetry.
+    Config for creating an evaluator: LLM provider configs, logger, telemetry.
 
     Logger defaults to the SDK package logger (``learning_commons_evaluators``), so
     log records propagate like typical library loggers. Pass ``logger=`` to use a
@@ -102,27 +100,27 @@ class EvaluatorConfig:
     create_config_telemetry_with_full_input, then pass the config to the evaluator constructor.
     """
 
-    google_prompt_provider_config: GooglePromptProviderConfig | None = None
-    openai_prompt_provider_config: OpenAIPromptProviderConfig | None = None
-    anthropic_prompt_provider_config: AnthropicPromptProviderConfig | None = None
+    google_llm_provider_config: GoogleLLMProviderConfig | None = None
+    openai_llm_provider_config: OpenAILLMProviderConfig | None = None
+    anthropic_llm_provider_config: AnthropicLLMProviderConfig | None = None
     logger: Logger = field(default_factory=get_logger)
     telemetry: TelemetryConfig = field(default_factory=TelemetryConfig)
 
 
 def create_config(
     *,
-    google_prompt_provider_config: GooglePromptProviderConfig | None = None,
-    openai_prompt_provider_config: OpenAIPromptProviderConfig | None = None,
-    anthropic_prompt_provider_config: AnthropicPromptProviderConfig | None = None,
+    google_llm_provider_config: GoogleLLMProviderConfig | None = None,
+    openai_llm_provider_config: OpenAILLMProviderConfig | None = None,
+    anthropic_llm_provider_config: AnthropicLLMProviderConfig | None = None,
     logger: Logger | None = None,
     telemetry_partner_id: str,
     send_full_input_with_telemetry: bool = False,
 ) -> EvaluatorConfig:
     """Create evaluator config with telemetry. telemetry_partner_id is required."""
     return EvaluatorConfig(
-        google_prompt_provider_config=google_prompt_provider_config,
-        openai_prompt_provider_config=openai_prompt_provider_config,
-        anthropic_prompt_provider_config=anthropic_prompt_provider_config,
+        google_llm_provider_config=google_llm_provider_config,
+        openai_llm_provider_config=openai_llm_provider_config,
+        anthropic_llm_provider_config=anthropic_llm_provider_config,
         logger=get_logger() if logger is None else logger,
         telemetry=TelemetryConfig(
             telemetry_partner_id=telemetry_partner_id,
@@ -133,16 +131,16 @@ def create_config(
 
 def create_config_no_telemetry(
     *,
-    google_prompt_provider_config: GooglePromptProviderConfig | None = None,
-    openai_prompt_provider_config: OpenAIPromptProviderConfig | None = None,
-    anthropic_prompt_provider_config: AnthropicPromptProviderConfig | None = None,
+    google_llm_provider_config: GoogleLLMProviderConfig | None = None,
+    openai_llm_provider_config: OpenAILLMProviderConfig | None = None,
+    anthropic_llm_provider_config: AnthropicLLMProviderConfig | None = None,
     logger: Logger | None = None,
 ) -> EvaluatorConfig:
     """Create evaluator config with telemetry disabled."""
     return EvaluatorConfig(
-        google_prompt_provider_config=google_prompt_provider_config,
-        openai_prompt_provider_config=openai_prompt_provider_config,
-        anthropic_prompt_provider_config=anthropic_prompt_provider_config,
+        google_llm_provider_config=google_llm_provider_config,
+        openai_llm_provider_config=openai_llm_provider_config,
+        anthropic_llm_provider_config=anthropic_llm_provider_config,
         logger=get_logger() if logger is None else logger,
         telemetry=TelemetryConfig(telemetry_partner_id=None, send_full_input_with_telemetry=False),
     )
@@ -150,17 +148,17 @@ def create_config_no_telemetry(
 
 def create_config_telemetry_with_full_input(
     *,
-    google_prompt_provider_config: GooglePromptProviderConfig | None = None,
-    openai_prompt_provider_config: OpenAIPromptProviderConfig | None = None,
-    anthropic_prompt_provider_config: AnthropicPromptProviderConfig | None = None,
+    google_llm_provider_config: GoogleLLMProviderConfig | None = None,
+    openai_llm_provider_config: OpenAILLMProviderConfig | None = None,
+    anthropic_llm_provider_config: AnthropicLLMProviderConfig | None = None,
     logger: Logger | None = None,
     telemetry_partner_id: str,
 ) -> EvaluatorConfig:
     """Create evaluator config with telemetry and full input sent with telemetry."""
     return EvaluatorConfig(
-        google_prompt_provider_config=google_prompt_provider_config,
-        openai_prompt_provider_config=openai_prompt_provider_config,
-        anthropic_prompt_provider_config=anthropic_prompt_provider_config,
+        google_llm_provider_config=google_llm_provider_config,
+        openai_llm_provider_config=openai_llm_provider_config,
+        anthropic_llm_provider_config=anthropic_llm_provider_config,
         logger=get_logger() if logger is None else logger,
         telemetry=TelemetryConfig(
             telemetry_partner_id=telemetry_partner_id, send_full_input_with_telemetry=True
diff --git a/sdks/python/src/learning_commons_evaluators/schemas/metadata.py b/sdks/python/src/learning_commons_evaluators/schemas/metadata.py
index 26de4dd..dd1c654 100644
--- a/sdks/python/src/learning_commons_evaluators/schemas/metadata.py
+++ b/sdks/python/src/learning_commons_evaluators/schemas/metadata.py
@@ -7,7 +7,7 @@
 from pydantic import BaseModel, Field, TypeAdapter, field_validator, model_validator
 
 from learning_commons_evaluators._version import __version__ as sdk_version
-from learning_commons_evaluators.schemas.config import LlmProvider, PromptSettings
+from learning_commons_evaluators.schemas.config import LLMProvider, PromptSettings
 from learning_commons_evaluators.schemas.input_specs import AnyInputSpec
 
 
@@ -79,7 +79,7 @@ def _normalize_maturity(cls, v: Any) -> Any:
 class TokenUsage(BaseModel):
     """Token usage for a some step of an evaluation: provider type, model, and token counts."""
 
-    provider_type: LlmProvider
+    provider_type: LLMProvider
     model: str
     input_tokens: int
     output_tokens: int
@@ -125,6 +125,6 @@ class EvaluationMetadata(BaseModel):
     input_metadata: InputMetadata
     status: Status = Status.processing
     error_details: str | None = None
-    total_token_usage: dict[LlmProvider, TokenUsage] = Field(default_factory=dict)
+    total_token_usage: dict[LLMProvider, TokenUsage] = Field(default_factory=dict)
     processing_time_ms: float = 0
     step_details: dict[str, StepMetadata] = Field(default_factory=dict)
diff --git a/sdks/python/src/learning_commons_evaluators/settings/_generated_conventionality_settings.py b/sdks/python/src/learning_commons_evaluators/settings/_generated_conventionality_settings.py
index a3c62b7..a83b9b1 100644
--- a/sdks/python/src/learning_commons_evaluators/settings/_generated_conventionality_settings.py
+++ b/sdks/python/src/learning_commons_evaluators/settings/_generated_conventionality_settings.py
@@ -5,7 +5,7 @@
 
 from __future__ import annotations
 
-from learning_commons_evaluators.schemas.config import LlmProvider, PromptSettings
+from learning_commons_evaluators.schemas.config import LLMProvider, PromptSettings
 from learning_commons_evaluators.schemas.conventionality import ConventionalityEvaluationSettings
 from learning_commons_evaluators.schemas.input_specs import GradeInputSpec, TextInputSpec
 from learning_commons_evaluators.schemas.metadata import EvaluatorMaturity, EvaluatorMetadata
@@ -86,7 +86,7 @@
 
 _EVALUATION_SETTINGS = ConventionalityEvaluationSettings(
     prompt_settings_step_conventionality_evaluation=PromptSettings(
-        provider_type=LlmProvider.GOOGLE,
+        provider_type=LLMProvider.GOOGLE,
         model='gemini-3-flash-preview',
         temperature=0.0,
     ),
diff --git a/sdks/python/src/learning_commons_evaluators/settings/_generated_vocabulary_settings.py b/sdks/python/src/learning_commons_evaluators/settings/_generated_vocabulary_settings.py
index 82562f8..5ad022c 100644
--- a/sdks/python/src/learning_commons_evaluators/settings/_generated_vocabulary_settings.py
+++ b/sdks/python/src/learning_commons_evaluators/settings/_generated_vocabulary_settings.py
@@ -5,7 +5,7 @@
 
 from __future__ import annotations
 
-from learning_commons_evaluators.schemas.config import LlmProvider, PromptSettings
+from learning_commons_evaluators.schemas.config import LLMProvider, PromptSettings
 from learning_commons_evaluators.schemas.input_specs import GradeInputSpec, TextInputSpec
 from learning_commons_evaluators.schemas.metadata import EvaluatorMaturity, EvaluatorMetadata
 from learning_commons_evaluators.schemas.vocabulary import VocabularyEvaluationSettings
@@ -383,16 +383,16 @@
 
 _EVALUATION_SETTINGS = VocabularyEvaluationSettings(
     prompt_settings_step_background_knowledge=PromptSettings(
-        provider_type=LlmProvider.OPENAI,
+        provider_type=LLMProvider.OPENAI,
         model='gpt-4o-2024-11-20',
         temperature=0.0,
     ),
     prompt_settings_step_vocab_grades_3_4=PromptSettings(
-        provider_type=LlmProvider.GOOGLE,
+        provider_type=LLMProvider.GOOGLE,
         model='gemini-2.5-pro',
         temperature=0.0,
     ),
-    prompt_settings_step_vocab_other_grades=PromptSettings(provider_type=LlmProvider.OPENAI, model='gpt-4.1', temperature=0.0),
+    prompt_settings_step_vocab_other_grades=PromptSettings(provider_type=LLMProvider.OPENAI, model='gpt-4.1', temperature=0.0),
 )
 
 # ── Public config object (imported by evaluator modules) ──────────────────────
diff --git a/sdks/python/tests/evaluators/test_base.py b/sdks/python/tests/evaluators/test_base.py
index 333e5c1..cc3f861 100644
--- a/sdks/python/tests/evaluators/test_base.py
+++ b/sdks/python/tests/evaluators/test_base.py
@@ -29,7 +29,7 @@
 from learning_commons_evaluators.schemas.common_inputs import GradeInputField, TextInputField
 from learning_commons_evaluators.schemas.config import (
     EvaluationSettings,
-    LlmProvider,
+    LLMProvider,
     PromptSettings,
 )
 from learning_commons_evaluators.schemas.errors import APIError, EvaluatorError, ValidationError
@@ -235,33 +235,33 @@ def emit(self, record: logging.LogRecord) -> None:
 class TestUpdateTotalTokenUsage:
     def test_inserts_usage_for_new_provider(self, stub_evaluator, evaluation_metadata):
         usage = TokenUsage(
-            provider_type=LlmProvider.GOOGLE,
+            provider_type=LLMProvider.GOOGLE,
             model="gemini-2.0-flash",
             input_tokens=10,
             output_tokens=5,
         )
         stub_evaluator.update_total_token_usage(usage, evaluation_metadata)
-        stored = evaluation_metadata.total_token_usage[LlmProvider.GOOGLE]
+        stored = evaluation_metadata.total_token_usage[LLMProvider.GOOGLE]
         assert stored.input_tokens == 10
         assert stored.output_tokens == 5
 
     def test_accumulates_usage_for_existing_provider(self, stub_evaluator, evaluation_metadata):
-        evaluation_metadata.total_token_usage[LlmProvider.GOOGLE] = TokenUsage(
-            provider_type=LlmProvider.GOOGLE,
+        evaluation_metadata.total_token_usage[LLMProvider.GOOGLE] = TokenUsage(
+            provider_type=LLMProvider.GOOGLE,
             model="gemini-2.0-flash",
             input_tokens=10,
             output_tokens=5,
         )
         stub_evaluator.update_total_token_usage(
             TokenUsage(
-                provider_type=LlmProvider.GOOGLE,
+                provider_type=LLMProvider.GOOGLE,
                 model="gemini-2.0-flash",
                 input_tokens=20,
                 output_tokens=15,
             ),
             evaluation_metadata,
         )
-        stored = evaluation_metadata.total_token_usage[LlmProvider.GOOGLE]
+        stored = evaluation_metadata.total_token_usage[LLMProvider.GOOGLE]
         assert stored.input_tokens == 30
         assert stored.output_tokens == 20
 
@@ -326,7 +326,7 @@ def _fake_llm(_pv):
             out = ev.execute_prompt_chain_step(
                 step_name="raw",
                 prompt_settings=PromptSettings(
-                    provider_type=LlmProvider.GOOGLE,
+                    provider_type=LLMProvider.GOOGLE,
                     model="gemini-2.0-flash",
                     temperature=0.0,
                 ),
@@ -345,7 +345,7 @@ def test_json_dict_normalizer_without_parser_type_raises(
             stub_evaluator.execute_prompt_chain_step(
                 step_name="raw",
                 prompt_settings=PromptSettings(
-                    provider_type=LlmProvider.GOOGLE,
+                    provider_type=LLMProvider.GOOGLE,
                     model="gemini-2.0-flash",
                     temperature=0.0,
                 ),
@@ -365,7 +365,7 @@ def _fake_llm(_pv):
             result = stub_evaluator.execute_prompt_chain_step(
                 step_name="main",
                 prompt_settings=PromptSettings(
-                    provider_type=LlmProvider.GOOGLE,
+                    provider_type=LLMProvider.GOOGLE,
                     model="gemini-2.0-flash",
                     temperature=0.0,
                 ),
@@ -400,7 +400,7 @@ def _double(d: dict) -> dict:
             result = stub_evaluator.execute_prompt_chain_step(
                 step_name="main",
                 prompt_settings=PromptSettings(
-                    provider_type=LlmProvider.GOOGLE,
+                    provider_type=LLMProvider.GOOGLE,
                     model="gemini-2.0-flash",
                     temperature=0.0,
                 ),
@@ -434,7 +434,7 @@ def _fake_llm(_pv):
             result = stub_evaluator.execute_prompt_chain_step(
                 step_name="main",
                 prompt_settings=PromptSettings(
-                    provider_type=LlmProvider.GOOGLE,
+                    provider_type=LLMProvider.GOOGLE,
                     model="gemini-2.0-flash",
                     temperature=0.0,
                 ),
@@ -461,7 +461,7 @@ def _fake_llm(_pv):
                 stub_evaluator.execute_prompt_chain_step(
                     step_name="main",
                     prompt_settings=PromptSettings(
-                        provider_type=LlmProvider.GOOGLE,
+                        provider_type=LLMProvider.GOOGLE,
                         model="gemini-2.0-flash",
                         temperature=0.0,
                     ),
@@ -487,7 +487,7 @@ def _fake_llm(_pv):
                 stub_evaluator.execute_prompt_chain_step(
                     step_name="main",
                     prompt_settings=PromptSettings(
-                        provider_type=LlmProvider.GOOGLE,
+                        provider_type=LLMProvider.GOOGLE,
                         model="gemini-2.0-flash",
                         temperature=0.0,
                     ),
@@ -500,7 +500,7 @@ def _fake_llm(_pv):
 
     def test_prompt_settings_recorded_in_step_extras(self, stub_evaluator, evaluation_metadata):
         settings = PromptSettings(
-            provider_type=LlmProvider.GOOGLE,
+            provider_type=LLMProvider.GOOGLE,
             model="gemini-2.0-flash",
             temperature=0.0,
         )
@@ -532,7 +532,7 @@ def _llm_with_usage(_pv):
             )
 
         settings = PromptSettings(
-            provider_type=LlmProvider.GOOGLE,
+            provider_type=LLMProvider.GOOGLE,
             model="gemini-2.0-flash",
             temperature=0.0,
         )
@@ -551,7 +551,7 @@ def _llm_with_usage(_pv):
         step = evaluation_metadata.step_details["main"]
         assert step.extras[PROMPT_STEP_EXTRA_TOKEN_USAGE]["input_tokens"] == 42
         assert step.extras[PROMPT_STEP_EXTRA_TOKEN_USAGE]["output_tokens"] == 17
-        assert evaluation_metadata.total_token_usage[LlmProvider.GOOGLE].input_tokens == 42
+        assert evaluation_metadata.total_token_usage[LLMProvider.GOOGLE].input_tokens == 42
 
     def test_propagates_configuration_error_from_create_provider(
         self, stub_evaluator, evaluation_metadata
@@ -567,7 +567,7 @@ def test_propagates_configuration_error_from_create_provider(
             stub_evaluator.execute_prompt_chain_step(
                 step_name="main",
                 prompt_settings=PromptSettings(
-                    provider_type=LlmProvider.GOOGLE,
+                    provider_type=LLMProvider.GOOGLE,
                     model="gemini-2.0-flash",
                     temperature=0.0,
                 ),
@@ -587,7 +587,7 @@ def test_propagates_evaluator_error_without_wrapping(self, stub_evaluator, evalu
             stub_evaluator.execute_prompt_chain_step(
                 step_name="main",
                 prompt_settings=PromptSettings(
-                    provider_type=LlmProvider.GOOGLE,
+                    provider_type=LLMProvider.GOOGLE,
                     model="gemini-2.0-flash",
                     temperature=0.0,
                 ),
@@ -609,7 +609,7 @@ def _boom(_pv):
             stub_evaluator.execute_prompt_chain_step(
                 step_name="main",
                 prompt_settings=PromptSettings(
-                    provider_type=LlmProvider.GOOGLE,
+                    provider_type=LLMProvider.GOOGLE,
                     model="gemini-2.0-flash",
                     temperature=0.0,
                 ),
@@ -633,7 +633,7 @@ def _bad(_pv):
             stub_evaluator.execute_prompt_chain_step(
                 step_name="main",
                 prompt_settings=PromptSettings(
-                    provider_type=LlmProvider.GOOGLE,
+                    provider_type=LLMProvider.GOOGLE,
                     model="gemini-2.0-flash",
                     temperature=0.0,
                 ),
@@ -659,7 +659,7 @@ def _partial(_pv):
             stub_evaluator.execute_prompt_chain_step(
                 step_name="main",
                 prompt_settings=PromptSettings(
-                    provider_type=LlmProvider.GOOGLE,
+                    provider_type=LLMProvider.GOOGLE,
                     model="gemini-2.0-flash",
                     temperature=0.0,
                 ),
diff --git a/sdks/python/tests/evaluators/test_conventionality.py b/sdks/python/tests/evaluators/test_conventionality.py
index e339da0..b3e4380 100644
--- a/sdks/python/tests/evaluators/test_conventionality.py
+++ b/sdks/python/tests/evaluators/test_conventionality.py
@@ -47,7 +47,7 @@ def test_evaluate_returns_evaluation_result(self):
 
     def test_evaluate_with_explicit_settings(self):
         from learning_commons_evaluators.schemas.config import (
-            LlmProvider,
+            LLMProvider,
             PromptSettings,
         )
         from learning_commons_evaluators.schemas.conventionality import (
@@ -58,7 +58,7 @@ def test_evaluate_with_explicit_settings(self):
         evaluator = ConventionalityEvaluator(config)
         settings = ConventionalityEvaluationSettings(
             prompt_settings_step_conventionality_evaluation=PromptSettings(
-                provider_type=LlmProvider.GOOGLE,
+                provider_type=LLMProvider.GOOGLE,
                 model="gemini-2.0-flash",
                 temperature=0.0,
             )
diff --git a/sdks/python/tests/providers/test_langchain.py b/sdks/python/tests/providers/test_langchain.py
index b9f9c31..0b0ff04 100644
--- a/sdks/python/tests/providers/test_langchain.py
+++ b/sdks/python/tests/providers/test_langchain.py
@@ -11,9 +11,9 @@
 )
 from learning_commons_evaluators.schemas.config import (
     EvaluatorConfig,
-    GooglePromptProviderConfig,
-    LlmProvider,
-    OpenAIPromptProviderConfig,
+    GoogleLLMProviderConfig,
+    LLMProvider,
+    OpenAILLMProviderConfig,
     PromptSettings,
 )
 
@@ -21,9 +21,9 @@
 def _config(**kwargs) -> EvaluatorConfig:
     """Return an EvaluatorConfig with all providers set to None; pass provider kwargs to override."""
     defaults: dict = {
-        "google_prompt_provider_config": None,
-        "openai_prompt_provider_config": None,
-        "anthropic_prompt_provider_config": None,
+        "google_llm_provider_config": None,
+        "openai_llm_provider_config": None,
+        "anthropic_llm_provider_config": None,
     }
     defaults.update(kwargs)
     return EvaluatorConfig(**defaults)
@@ -36,38 +36,34 @@ def _config(**kwargs) -> EvaluatorConfig:
 
 class TestCreateProvider:
     def test_google_provider_returns_model(self):
-        config = _config(
-            google_prompt_provider_config=GooglePromptProviderConfig(api_key="test-key")
-        )
+        config = _config(google_llm_provider_config=GoogleLLMProviderConfig(api_key="test-key"))
         settings = PromptSettings(
-            provider_type=LlmProvider.GOOGLE, model="gemini-2.0-flash", temperature=0.0
+            provider_type=LLMProvider.GOOGLE, model="gemini-2.0-flash", temperature=0.0
         )
         assert create_provider(settings, config) is not None
 
     def test_openai_provider_returns_model(self):
-        config = _config(
-            openai_prompt_provider_config=OpenAIPromptProviderConfig(api_key="test-key")
-        )
+        config = _config(openai_llm_provider_config=OpenAILLMProviderConfig(api_key="test-key"))
         settings = PromptSettings(
-            provider_type=LlmProvider.OPENAI, model="gpt-4o-mini", temperature=0.0
+            provider_type=LLMProvider.OPENAI, model="gpt-4o-mini", temperature=0.0
         )
         assert create_provider(settings, config) is not None
 
     def test_raises_when_google_config_missing(self):
         settings = PromptSettings(
-            provider_type=LlmProvider.GOOGLE, model="gemini-2.0-flash", temperature=0.0
+            provider_type=LLMProvider.GOOGLE, model="gemini-2.0-flash", temperature=0.0
         )
         with pytest.raises(ConfigurationError, match="Google provider config is not set"):
             create_provider(settings, _config())
 
     def test_raises_when_openai_config_missing(self):
-        settings = PromptSettings(provider_type=LlmProvider.OPENAI, model="gpt-4o", temperature=0.0)
+        settings = PromptSettings(provider_type=LLMProvider.OPENAI, model="gpt-4o", temperature=0.0)
         with pytest.raises(ConfigurationError, match="OpenAI provider config is not set"):
             create_provider(settings, _config())
 
     def test_raises_when_anthropic_config_missing(self):
         settings = PromptSettings(
-            provider_type=LlmProvider.ANTHROPIC, model="claude-3", temperature=0.0
+            provider_type=LLMProvider.ANTHROPIC, model="claude-3", temperature=0.0
         )
         with pytest.raises(ConfigurationError, match="Anthropic provider config is not set"):
             create_provider(settings, _config())
@@ -79,7 +75,7 @@ def test_raises_configuration_error_for_unrecognized_provider_type(self):
         with pytest.raises(ConfigurationError, match="Unsupported LLM provider type"):
             create_provider(
                 mock_settings,
-                _config(google_prompt_provider_config=GooglePromptProviderConfig(api_key="k")),
+                _config(google_llm_provider_config=GoogleLLMProviderConfig(api_key="k")),
             )
 
 
@@ -91,23 +87,23 @@ def test_raises_configuration_error_for_unrecognized_provider_type(self):
 class TestTokenUsageFromAIMessage:
     def test_returns_zero_usage_when_no_usage_metadata(self):
         settings = PromptSettings(
-            provider_type=LlmProvider.GOOGLE, model="gemini-2.0-flash", temperature=0.0
+            provider_type=LLMProvider.GOOGLE, model="gemini-2.0-flash", temperature=0.0
         )
         usage = token_usage_from_aimessage(object(), settings)
-        assert usage.provider_type == LlmProvider.GOOGLE
+        assert usage.provider_type == LLMProvider.GOOGLE
         assert usage.model == "gemini-2.0-flash"
         assert usage.input_tokens == 0
         assert usage.output_tokens == 0
 
     def test_uses_usage_metadata_when_present(self):
-        settings = PromptSettings(provider_type=LlmProvider.OPENAI, model="gpt-4o", temperature=0.0)
+        settings = PromptSettings(provider_type=LLMProvider.OPENAI, model="gpt-4o", temperature=0.0)
         message = type("Msg", (), {"usage_metadata": {"input_tokens": 100, "output_tokens": 50}})()
         usage = token_usage_from_aimessage(message, settings)
         assert usage.input_tokens == 100
         assert usage.output_tokens == 50
 
     def test_falls_back_to_response_metadata_when_usage_metadata_absent(self):
-        settings = PromptSettings(provider_type=LlmProvider.GOOGLE, model="gemini", temperature=0.0)
+        settings = PromptSettings(provider_type=LLMProvider.GOOGLE, model="gemini", temperature=0.0)
         message = type(
             "Msg",
             (),
diff --git a/sdks/python/tests/schemas/test_config.py b/sdks/python/tests/schemas/test_config.py
index 6a50b09..92942f4 100644
--- a/sdks/python/tests/schemas/test_config.py
+++ b/sdks/python/tests/schemas/test_config.py
@@ -1,4 +1,4 @@
-"""Tests for EvaluatorConfig, PromptProviderConfig subclasses, and factory functions."""
+"""Tests for EvaluatorConfig, LLMProviderConfig subclasses, and factory functions."""
 
 from dataclasses import FrozenInstanceError
 
@@ -6,10 +6,10 @@
 
 from learning_commons_evaluators.logger import SDK_LOGGER_NAME, get_logger
 from learning_commons_evaluators.schemas.config import (
-    AnthropicPromptProviderConfig,
-    GooglePromptProviderConfig,
-    LlmProvider,
-    OpenAIPromptProviderConfig,
+    AnthropicLLMProviderConfig,
+    GoogleLLMProviderConfig,
+    LLMProvider,
+    OpenAILLMProviderConfig,
     TelemetryConfig,
     create_config,
     create_config_no_telemetry,
@@ -17,26 +17,26 @@
 )
 
 
-class TestLlmProvider:
+class TestLLMProvider:
     @pytest.mark.parametrize(
         "member,value",
         [
-            (LlmProvider.ANTHROPIC, "anthropic"),
-            (LlmProvider.GOOGLE, "google"),
-            (LlmProvider.OPENAI, "openai"),
+            (LLMProvider.ANTHROPIC, "anthropic"),
+            (LLMProvider.GOOGLE, "google"),
+            (LLMProvider.OPENAI, "openai"),
         ],
     )
     def test_provider_value(self, member, value):
         assert member.value == value
 
 
-class TestPromptProviderConfigs:
+class TestLLMProviderConfigs:
     @pytest.mark.parametrize(
         "cls,expected_type",
         [
-            (GooglePromptProviderConfig, LlmProvider.GOOGLE),
-            (OpenAIPromptProviderConfig, LlmProvider.OPENAI),
-            (AnthropicPromptProviderConfig, LlmProvider.ANTHROPIC),
+            (GoogleLLMProviderConfig, LLMProvider.GOOGLE),
+            (OpenAILLMProviderConfig, LLMProvider.OPENAI),
+            (AnthropicLLMProviderConfig, LLMProvider.ANTHROPIC),
         ],
     )
     def test_provider_config_default_type(self, cls, expected_type):
@@ -54,11 +54,11 @@ def test_create_config_no_telemetry_defaults(self):
 
     def test_create_config_no_telemetry_accepts_providers(self):
         config = create_config_no_telemetry(
-            google_prompt_provider_config=GooglePromptProviderConfig(api_key="gk"),
-            openai_prompt_provider_config=OpenAIPromptProviderConfig(api_key="ok"),
+            google_llm_provider_config=GoogleLLMProviderConfig(api_key="gk"),
+            openai_llm_provider_config=OpenAILLMProviderConfig(api_key="ok"),
         )
-        assert config.google_prompt_provider_config.api_key == "gk"
-        assert config.openai_prompt_provider_config.api_key == "ok"
+        assert config.google_llm_provider_config.api_key == "gk"
+        assert config.openai_llm_provider_config.api_key == "ok"
 
     def test_create_config_sets_telemetry_partner_id(self):
         config = create_config(telemetry_partner_id="tid-123")
diff --git a/sdks/python/tests/schemas/test_metadata.py b/sdks/python/tests/schemas/test_metadata.py
index 6ddbd81..67e9c35 100644
--- a/sdks/python/tests/schemas/test_metadata.py
+++ b/sdks/python/tests/schemas/test_metadata.py
@@ -9,7 +9,7 @@
 import pytest
 from pydantic import ValidationError
 
-from learning_commons_evaluators.schemas.config import LlmProvider, PromptSettings
+from learning_commons_evaluators.schemas.config import LLMProvider, PromptSettings
 from learning_commons_evaluators.schemas.input_specs import TextInputSpec
 from learning_commons_evaluators.schemas.metadata import (
     PROMPT_STEP_EXTRA_PROMPT_SETTINGS,
@@ -98,18 +98,18 @@ def test_model_validate_unknown_input_type_raises(self):
 
     def test_fields(self):
         usage = TokenUsage(
-            provider_type=LlmProvider.GOOGLE,
+            provider_type=LLMProvider.GOOGLE,
             model="gemini-2.0-flash",
             input_tokens=100,
             output_tokens=50,
         )
-        assert usage.provider_type == LlmProvider.GOOGLE
+        assert usage.provider_type == LLMProvider.GOOGLE
         assert usage.input_tokens == 100
         assert usage.output_tokens == 50
 
     def test_zero_tokens_are_valid(self):
         usage = TokenUsage(
-            provider_type=LlmProvider.OPENAI,
+            provider_type=LLMProvider.OPENAI,
             model="gpt-4o-mini",
             input_tokens=0,
             output_tokens=0,
@@ -139,7 +139,7 @@ class TestPromptSettingsToExtrasValue:
     def test_produces_json_serialisable_dict(self):
         """provider_type must be a plain string (not the enum) so the dict is JSON-safe."""
         settings = PromptSettings(
-            provider_type=LlmProvider.ANTHROPIC,
+            provider_type=LLMProvider.ANTHROPIC,
             model="claude-3-haiku",
             temperature=0.5,
         )
diff --git a/sdks/python/tests/settings/test_load_settings.py b/sdks/python/tests/settings/test_load_settings.py
index 25dc88b..8fd3882 100644
--- a/sdks/python/tests/settings/test_load_settings.py
+++ b/sdks/python/tests/settings/test_load_settings.py
@@ -12,7 +12,7 @@
 
 from learning_commons_evaluators.schemas.config import (
     EvaluationSettings,
-    LlmProvider,
+    LLMProvider,
     PromptSettings,
 )
 from learning_commons_evaluators.schemas.conventionality import (
@@ -200,7 +200,7 @@ def test_load_evaluator_settings_prompt_settings_and_prompts(tmp_path: Path) ->
         ).strip()
     )
     result = load_evaluator_settings(path, _WithPromptSettings)
-    assert result.evaluation_settings.prompt_settings_main.provider_type == LlmProvider.GOOGLE
+    assert result.evaluation_settings.prompt_settings_main.provider_type == LLMProvider.GOOGLE
     assert result.evaluation_settings.prompt_settings_main.model == "gemini-2.0-flash"
     assert result.evaluation_settings.prompt_settings_main.temperature == 0.25
     assert result.prompts["system_prompt"] == "hello\n\nworld"