yera.config.schema.models

Model schema — universe of all discovered models and runtime catalog view.

Symbols

class AnthropicLLMCapabilities — Capabilities specific to Anthropic models.
class AnthropicLLMInference — Anthropic inference settings model.
class AwsLLMCapabilities — Capabilities of AWS Bedrock LLMs.
class AwsLLMInference — AWS Bedrock inference settings model.
class BaseModelConfig — Fields common to all model types.
class EmbeddingsConfig — Config for an embeddings model.
class GeminiLLMCapabilities — Capabilities reported or looked up for a Gemini model.
class GeminiLLMInference — Inference settings for Gemini Developer API models.
class GenericLLMCapabilities — Capabilities for a generic LLM (no provider-specific features).
class GenericLLMInference — Default inference settings representation.
class LlamaCppLLMCapabilities — Capabilities of llama-cpp LLMs.
class LlamaCppLLMInference — Llama-cpp inference settings model.
class LLMCapabilities — Base for all LLM capabilities models.
class LLMConfig — Config for a large language model.
class LLMInference — Base class for all LLM inference settings models.
class MistralLLMCapabilities — Capabilities of Mistral LLMs.
class MistralLLMInference — Mistral inference settings model.
class ModelCatalogue — Runtime view of the model universe filtered to the active Yera profile.
class ModelsUniverse — All discovered models across all provider connections.
class OpenAILLMCapabilities — Capabilities of OpenAI LLMs.
class OpenAILLMInference — OpenAI inference settings model.
class OpenRouterLLMCapabilities — Capabilities reported for an OpenRouter model.
class OpenRouterLLMInference — Inference settings for models invoked through OpenRouter.
class RerankerConfig — Config for a reranker model.
class STTConfig — Config for a speech-to-text model.
class TTSConfig — Config for a text-to-speech model.
class TypedCatalogue — A filtered runtime view of models of one type for the active Yera profile.

AnthropicLLMCapabilities

Inherits: LLMCapabilities

Capabilities specific to Anthropic models.

Fields

supports_vision
type: bool = False

Whether the model accepts image input.

supports_pdf
type: bool = False

Whether the model accepts PDF document input.

reasoning
type: Literal['effort', 'budget', 'none', 'always'] = 'none'

The reasoning control supported by the model.

reasoning_levels
type: tuple[typing.Literal['off', 'minimal', 'low', 'medium', 'high', 'xhigh', 'max', 'auto'], ...] = ()

Reasoning levels accepted by the model.

reasoning_mandatory
type: bool = False

Whether reasoning must remain enabled for the model.

provider
type: Literal['anthropic'] = 'anthropic'

Discriminator for Anthropic capability settings.

structured_outputs
type: bool = True

Whether the model supports native structured output.

mid_conversation_system
type: bool = False

Whether system instructions can appear after the conversation has begun.

AnthropicLLMInference

Inherits: LLMInference

Anthropic inference settings model.

Fields

max_tokens
type: int | None = None

Default maximum number of output tokens.

temperature
type: float | None = None

Default sampling temperature.

top_p
type: float | None = None

Default nucleus-sampling probability.

reasoning_effort
type: Optional[Literal['off', 'minimal', 'low', 'medium', 'high', 'xhigh', 'max', 'auto']] = None

Default reasoning level when supported by the model.

reasoning_budget
type: int | None = None

Default token budget reserved for reasoning.

provider
type: Literal['anthropic'] = 'anthropic'

Discriminator for Anthropic inference settings.

display
type: Literal['summarized', 'omitted'] = 'summarized'

How reasoning output is represented in responses.

effort
type: Optional[Literal['low', 'medium', 'high', 'xhigh', 'max']] = None

Default Anthropic reasoning effort.

cache
type: bool = False

Whether prompt caching is enabled by default.

cache_ttl
type: Literal['5m', '1h'] = '5m'

Default lifetime of cached prompt content.

AwsLLMCapabilities

Inherits: LLMCapabilities

Capabilities of AWS Bedrock LLMs.

Fields

supports_vision
type: bool = False

Whether the model accepts image input.

supports_pdf
type: bool = False

Whether the model accepts PDF document input.

reasoning
type: Literal['effort', 'budget', 'none', 'always'] = 'none'

The reasoning control supported by the model.

reasoning_levels
type: tuple[typing.Literal['off', 'minimal', 'low', 'medium', 'high', 'xhigh', 'max', 'auto'], ...] = ()

Reasoning levels accepted by the model.

reasoning_mandatory
type: bool = False

Whether reasoning must remain enabled for the model.

provider
type: Literal['aws'] = 'aws'

Discriminator for AWS Bedrock capability settings.

transport
type: Literal['converse', 'responses'] = 'converse'

AWS Bedrock API used to invoke the model.

AwsLLMInference

Inherits: LLMInference

AWS Bedrock inference settings model.

Fields

max_tokens
type: int | None = None

Default maximum number of output tokens.

temperature
type: float | None = None

Default sampling temperature.

top_p
type: float | None = None

Default nucleus-sampling probability.

reasoning_effort
type: Optional[Literal['off', 'minimal', 'low', 'medium', 'high', 'xhigh', 'max', 'auto']] = None

Default reasoning level when supported by the model.

reasoning_budget
type: int | None = None

Default token budget reserved for reasoning.

provider
type: Literal['aws'] = 'aws'

Discriminator for AWS Bedrock inference settings.

BaseModelConfig

Inherits: BaseModel

Fields common to all model types.

Raises

ValueError

If id has fewer than three dot-delimited segments.

Fields

id
type: str (required)

Yera-internal dot-delimited identifier. Must have at least three segments: <provider>.<author>.<model_name>, with optional further nesting for variants e.g. llama_cpp.meta.llama3.8b.q4_k_m. Does not encode the connection name — that is stored separately in connection.

model_id
type: str (required)

Provider-side model identifier passed directly to the API or library, e.g. "claude-sonnet-4-20250514" or a HuggingFace repo id like "meta-llama/Meta-Llama-3-8B-GGUF".

name
type: str (required)

Human-readable display name, e.g. "Claude Sonnet 4".

provider
type: str (required)

Key into :class:~yera.config.schema.providers.Providers identifying which provider serves this model, e.g. "aws".

connection
type: str (required)

Name of the provider connection under which this model was discovered, e.g. "work". Together with provider, uniquely identifies the runtime interface and credentials to use.

author
type: str (required)

Who created the model, e.g. "anthropic" or "meta". Distinct from provider — a Meta model can be served via AWS Bedrock.

Methods

id_has_minimum_depth — Check that id has at least three dot-delimited segments.
id_prefix_matches_provider — Check that the first segment of id matches provider.

BaseModelConfig.id_has_minimum_depth

id_has_minimum_depth(
    v: str,
) → str

Check that id has at least three dot-delimited segments.

Parameters

v
type: str

The id string to validate.

Returns

type: str

The validated id string.

Raises

ValueError

If the id has fewer than three segments.

BaseModelConfig.id_prefix_matches_provider

id_prefix_matches_provider() → BaseModelConfig

Check that the first segment of id matches provider.

Returns

type: BaseModelConfig

The validated model config.

Raises

ValueError

If the first segment of id does not match provider.

EmbeddingsConfig

Inherits: BaseModelConfig

Config for an embeddings model.

Fields

id
type: str (required)

Yera-internal dot-delimited identifier. Must have at least three segments: <provider>.<author>.<model_name>, with optional further nesting for variants e.g. llama_cpp.meta.llama3.8b.q4_k_m. Does not encode the connection name — that is stored separately in connection.

model_id
type: str (required)

Provider-side model identifier passed directly to the API or library, e.g. "claude-sonnet-4-20250514" or a HuggingFace repo id like "meta-llama/Meta-Llama-3-8B-GGUF".

name
type: str (required)

Human-readable display name, e.g. "Claude Sonnet 4".

provider
type: str (required)

Key into :class:~yera.config.schema.providers.Providers identifying which provider serves this model, e.g. "aws".

connection
type: str (required)

Name of the provider connection under which this model was discovered, e.g. "work". Together with provider, uniquely identifies the runtime interface and credentials to use.

author
type: str (required)

Who created the model, e.g. "anthropic" or "meta". Distinct from provider — a Meta model can be served via AWS Bedrock.

type
type: Literal['embedding'] = 'embedding'

Discriminator field, always "embedding".

dimensions
type: int | None = None

Size of the embedding vector produced by the model.

max_input_tokens
type: int | None = None

Maximum number of tokens the model can embed in a single request.

GeminiLLMCapabilities

Inherits: LLMCapabilities

Capabilities reported or looked up for a Gemini model.

Fields

supports_vision
type: bool = False

Whether the model accepts image input.

supports_pdf
type: bool = False

Whether the model accepts PDF document input.

reasoning
type: Literal['effort', 'budget', 'none', 'always'] = 'none'

The reasoning control supported by the model.

reasoning_levels
type: tuple[typing.Literal['off', 'minimal', 'low', 'medium', 'high', 'xhigh', 'max', 'auto'], ...] = ()

Reasoning levels accepted by the model.

reasoning_mandatory
type: bool = False

Whether reasoning must remain enabled for the model.

provider
type: Literal['gemini'] = 'gemini'

Discriminator for Gemini capability settings.

structured_outputs
type: bool = False

Whether the model supports native structured output.

tools
type: bool = False

Whether the model supports tool calls.

tool_choice
type: bool = False

Whether requests can control how the model selects tools.

GeminiLLMInference

Inherits: LLMInference

Inference settings for Gemini Developer API models.

Fields

max_tokens
type: int | None = None

Default maximum number of output tokens.

temperature
type: float | None = None

Default sampling temperature.

top_p
type: float | None = None

Default nucleus-sampling probability.

reasoning_effort
type: Optional[Literal['off', 'minimal', 'low', 'medium', 'high', 'xhigh', 'max', 'auto']] = None

Default reasoning level when supported by the model.

reasoning_budget
type: int | None = None

Default token budget reserved for reasoning.

provider
type: Literal['gemini'] = 'gemini'

Discriminator for Gemini inference settings.

top_k
type: int | None = None

Default number of highest-probability tokens considered.

include_thoughts
type: bool = True

Whether thought summaries are included by default.

GenericLLMCapabilities

Inherits: LLMCapabilities

Capabilities for a generic LLM (no provider-specific features).

Fields

supports_vision
type: bool = False

Whether the model accepts image input.

supports_pdf
type: bool = False

Whether the model accepts PDF document input.

reasoning
type: Literal['effort', 'budget', 'none', 'always'] = 'none'

The reasoning control supported by the model.

reasoning_levels
type: tuple[typing.Literal['off', 'minimal', 'low', 'medium', 'high', 'xhigh', 'max', 'auto'], ...] = ()

Reasoning levels accepted by the model.

reasoning_mandatory
type: bool = False

Whether reasoning must remain enabled for the model.

provider
type: Literal['generic'] = 'generic'

Discriminator for capabilities without provider-specific behavior.

GenericLLMInference

Inherits: LLMInference

Default inference settings representation.

Fields

max_tokens
type: int | None = None

Default maximum number of output tokens.

temperature
type: float | None = None

Default sampling temperature.

top_p
type: float | None = None

Default nucleus-sampling probability.

reasoning_effort
type: Optional[Literal['off', 'minimal', 'low', 'medium', 'high', 'xhigh', 'max', 'auto']] = None

Default reasoning level when supported by the model.

reasoning_budget
type: int | None = None

Default token budget reserved for reasoning.

provider
type: Literal['generic'] = 'generic'

Discriminator for inference settings without provider-specific behavior.

LlamaCppLLMCapabilities

Inherits: LLMCapabilities

Capabilities of llama-cpp LLMs.

Fields

supports_vision
type: bool = False

Whether the model accepts image input.

supports_pdf
type: bool = False

Whether the model accepts PDF document input.

reasoning
type: Literal['effort', 'budget', 'none', 'always'] = 'none'

The reasoning control supported by the model.

reasoning_levels
type: tuple[typing.Literal['off', 'minimal', 'low', 'medium', 'high', 'xhigh', 'max', 'auto'], ...] = ()

Reasoning levels accepted by the model.

reasoning_mandatory
type: bool = False

Whether reasoning must remain enabled for the model.

provider
type: Literal['llama_cpp'] = 'llama_cpp'

Discriminator for llama.cpp capability settings.

structured_outputs
type: bool = True

Whether the model supports structured output through llama.cpp.

LlamaCppLLMInference

Inherits: LLMInference

Llama-cpp inference settings model.

Fields

max_tokens
type: int | None = None

Default maximum number of output tokens.

temperature
type: float | None = None

Default sampling temperature.

top_p
type: float | None = None

Default nucleus-sampling probability.

reasoning_effort
type: Optional[Literal['off', 'minimal', 'low', 'medium', 'high', 'xhigh', 'max', 'auto']] = None

Default reasoning level when supported by the model.

reasoning_budget
type: int | None = None

Default token budget reserved for reasoning.

provider
type: Literal['llama_cpp'] = 'llama_cpp'

Discriminator for llama.cpp inference settings.

n_gpu_layers
type: int = -1

Number of model layers assigned to the GPU.

context_length
type: int = 0

Default llama-cpp context length.

LLMCapabilities

Inherits: BaseModel

Base for all LLM capabilities models.

Fields

supports_vision
type: bool = False

Whether the model accepts image input.

supports_pdf
type: bool = False

Whether the model accepts PDF document input.

reasoning
type: Literal['effort', 'budget', 'none', 'always'] = 'none'

The reasoning control supported by the model.

reasoning_levels
type: tuple[typing.Literal['off', 'minimal', 'low', 'medium', 'high', 'xhigh', 'max', 'auto'], ...] = ()

Reasoning levels accepted by the model.

reasoning_mandatory
type: bool = False

Whether reasoning must remain enabled for the model.

LLMConfig

Inherits: BaseModelConfig

Config for a large language model.

Fields

id
type: str (required)

Yera-internal dot-delimited identifier. Must have at least three segments: <provider>.<author>.<model_name>, with optional further nesting for variants e.g. llama_cpp.meta.llama3.8b.q4_k_m. Does not encode the connection name — that is stored separately in connection.

model_id
type: str (required)

Provider-side model identifier passed directly to the API or library, e.g. "claude-sonnet-4-20250514" or a HuggingFace repo id like "meta-llama/Meta-Llama-3-8B-GGUF".

name
type: str (required)

Human-readable display name, e.g. "Claude Sonnet 4".

provider
type: str (required)

Key into :class:~yera.config.schema.providers.Providers identifying which provider serves this model, e.g. "aws".

connection
type: str (required)

Name of the provider connection under which this model was discovered, e.g. "work". Together with provider, uniquely identifies the runtime interface and credentials to use.

author
type: str (required)

Who created the model, e.g. "anthropic" or "meta". Distinct from provider — a Meta model can be served via AWS Bedrock.

type
type: Literal['llm'] = 'llm'

Discriminator field, always "llm".

context_window
type: int | None = None

Maximum number of tokens the model can process in a single request (prompt + completion).

max_output_tokens
type: int | None = None

Maximum number of tokens the model can generate.

capabilities
type: Union[Annotated[yera.config.schema.models.GenericLLMCapabilities, Tag(tag='generic')], Annotated[yera.config.schema.models.AnthropicLLMCapabilities, Tag(tag='anthropic')], Annotated[yera.config.schema.models.LlamaCppLLMCapabilities, Tag(tag='llama_cpp')], Annotated[yera.config.schema.models.MistralLLMCapabilities, Tag(tag='mistral')], Annotated[yera.config.schema.models.OpenAILLMCapabilities, Tag(tag='openai')], Annotated[yera.config.schema.models.AwsLLMCapabilities, Tag(tag='aws')], Annotated[yera.config.schema.models.OpenRouterLLMCapabilities, Tag(tag='openrouter')], Annotated[yera.config.schema.models.GeminiLLMCapabilities, Tag(tag='gemini')]] = <factory>

Features and input modes supported by the model.

inference
type: Union[Annotated[yera.config.schema.models.GenericLLMInference, Tag(tag='generic')], Annotated[yera.config.schema.models.AnthropicLLMInference, Tag(tag='anthropic')], Annotated[yera.config.schema.models.LlamaCppLLMInference, Tag(tag='llama_cpp')], Annotated[yera.config.schema.models.MistralLLMInference, Tag(tag='mistral')], Annotated[yera.config.schema.models.OpenAILLMInference, Tag(tag='openai')], Annotated[yera.config.schema.models.AwsLLMInference, Tag(tag='aws')], Annotated[yera.config.schema.models.GeminiLLMInference, Tag(tag='gemini')], Annotated[yera.config.schema.models.OpenRouterLLMInference, Tag(tag='openrouter')]] = GenericLLMInference(max_tokens=None, temperature=None, top_p=None, reasoning_effort=None, reasoning_budget=None, provider='generic')

Default provider-specific settings used when invoking the model.

LLMInference

Inherits: BaseModel

Base class for all LLM inference settings models.

Fields

max_tokens
type: int | None = None

Default maximum number of output tokens.

temperature
type: float | None = None

Default sampling temperature.

top_p
type: float | None = None

Default nucleus-sampling probability.

reasoning_effort
type: Optional[Literal['off', 'minimal', 'low', 'medium', 'high', 'xhigh', 'max', 'auto']] = None

Default reasoning level when supported by the model.

reasoning_budget
type: int | None = None

Default token budget reserved for reasoning.

MistralLLMCapabilities

Inherits: LLMCapabilities

Capabilities of Mistral LLMs.

Fields

supports_vision
type: bool = False

Whether the model accepts image input.

supports_pdf
type: bool = False

Whether the model accepts PDF document input.

reasoning
type: Literal['effort', 'budget', 'none', 'always'] = 'none'

The reasoning control supported by the model.

reasoning_levels
type: tuple[typing.Literal['off', 'minimal', 'low', 'medium', 'high', 'xhigh', 'max', 'auto'], ...] = ()

Reasoning levels accepted by the model.

reasoning_mandatory
type: bool = False

Whether reasoning must remain enabled for the model.

provider
type: Literal['mistral'] = 'mistral'

Discriminator for Mistral capability settings.

MistralLLMInference

Inherits: LLMInference

Mistral inference settings model.

Fields

max_tokens
type: int | None = None

Default maximum number of output tokens.

temperature
type: float | None = None

Default sampling temperature.

top_p
type: float | None = None

Default nucleus-sampling probability.

reasoning_effort
type: Optional[Literal['off', 'minimal', 'low', 'medium', 'high', 'xhigh', 'max', 'auto']] = None

Default reasoning level when supported by the model.

reasoning_budget
type: int | None = None

Default token budget reserved for reasoning.

provider
type: Literal['mistral'] = 'mistral'

Discriminator for Mistral inference settings.

ModelCatalogue

Inherits: BaseModel

Runtime view of the model universe filtered to the active Yera profile.

Never serialized — constructed by :meth:ModelsUniverse.for_profile. Contains only models whose provider and connection match the active Yera profile's provider map.

Fields

llm
type: yera.config.schema.models.TypedCatalogue[LLMConfig] (required)

Catalogue of large language models visible under the active profile.

embedding
type: yera.config.schema.models.TypedCatalogue[EmbeddingsConfig] (required)

Catalogue of embeddings models visible under the active profile.

tts
type: yera.config.schema.models.TypedCatalogue[TTSConfig] (required)

Catalogue of text-to-speech models visible under the active profile.

stt
type: yera.config.schema.models.TypedCatalogue[STTConfig] (required)

Catalogue of speech-to-text models visible under the active profile.

reranker
type: yera.config.schema.models.TypedCatalogue[RerankerConfig] (required)

Catalogue of reranker models visible under the active profile.

ModelsUniverse

Inherits: BaseModel

All discovered models across all provider connections.

Serialized to and from yera.toml under [models]. Contains no defaults — those live in :class:~yera.config.schema.profiles.Profile.

Fields

llm
type: list[yera.config.schema.models.LLMConfig] = <factory>

All discovered large language models.

embedding
type: list[yera.config.schema.models.EmbeddingsConfig] = <factory>

All discovered embeddings models.

tts
type: list[yera.config.schema.models.TTSConfig] = <factory>

All discovered text-to-speech models.

stt
type: list[yera.config.schema.models.STTConfig] = <factory>

All discovered speech-to-text models.

reranker
type: list[yera.config.schema.models.RerankerConfig] = <factory>

All discovered reranker models.

Methods

for_profile — Build a filtered runtime ``ModelCatalogue`` for the active Yera profile.

ModelsUniverse.for_profile

for_profile(
    profile: Profile,
) → ModelCatalogue

Build a filtered runtime ModelCatalogue for the active Yera profile.

Filters each model type list to models whose provider and connection match an entry in the profile's provider map, then inserts per-type defaults from the profile's model_defaults.

Parameters

profile
type: Profile

The active Yera profile to filter and resolve against.

Returns

type: ModelCatalogue

A ModelCatalogue containing only models visible under the active profile, with defaults populated from the profile.

OpenAILLMCapabilities

Inherits: LLMCapabilities

Capabilities of OpenAI LLMs.

Fields

supports_vision
type: bool = False

Whether the model accepts image input.

supports_pdf
type: bool = False

Whether the model accepts PDF document input.

reasoning
type: Literal['effort', 'budget', 'none', 'always'] = 'none'

The reasoning control supported by the model.

reasoning_levels
type: tuple[typing.Literal['off', 'minimal', 'low', 'medium', 'high', 'xhigh', 'max', 'auto'], ...] = ()

Reasoning levels accepted by the model.

reasoning_mandatory
type: bool = False

Whether reasoning must remain enabled for the model.

provider
type: Literal['openai'] = 'openai'

Discriminator for OpenAI capability settings.

OpenAILLMInference

Inherits: LLMInference

OpenAI inference settings model.

Fields

max_tokens
type: int | None = None

Default maximum number of output tokens.

temperature
type: float | None = None

Default sampling temperature.

top_p
type: float | None = None

Default nucleus-sampling probability.

reasoning_effort
type: Optional[Literal['off', 'minimal', 'low', 'medium', 'high', 'xhigh', 'max', 'auto']] = None

Default reasoning level when supported by the model.

reasoning_budget
type: int | None = None

Default token budget reserved for reasoning.

provider
type: Literal['openai'] = 'openai'

Discriminator for OpenAI inference settings.

summary
type: Optional[Literal['auto', 'concise', 'detailed']] = 'auto'

Default level of detail in reasoning summaries.

OpenRouterLLMCapabilities

Inherits: LLMCapabilities

Capabilities reported for an OpenRouter model.

Fields

supports_vision
type: bool = False

Whether the model accepts image input.

supports_pdf
type: bool = False

Whether the model accepts PDF document input.

reasoning
type: Literal['effort', 'budget', 'none', 'always'] = 'none'

The reasoning control supported by the model.

reasoning_levels
type: tuple[typing.Literal['off', 'minimal', 'low', 'medium', 'high', 'xhigh', 'max', 'auto'], ...] = ()

Reasoning levels accepted by the model.

reasoning_mandatory
type: bool = False

Whether reasoning must remain enabled for the model.

provider
type: Literal['openrouter'] = 'openrouter'

Discriminator for OpenRouter capability settings.

structured_outputs
type: bool = False

Whether the model supports native structured output.

tools
type: bool = False

Whether the model supports tool calls.

tool_choice
type: bool = False

Whether requests can control how the model selects tools.

supports_reasoning_budget
type: bool = False

Whether requests can set an explicit reasoning-token budget.

OpenRouterLLMInference

Inherits: LLMInference

Inference settings for models invoked through OpenRouter.

Fields

max_tokens
type: int | None = None

Default maximum number of output tokens.

temperature
type: float | None = None

Default sampling temperature.

top_p
type: float | None = None

Default nucleus-sampling probability.

reasoning_effort
type: Optional[Literal['off', 'minimal', 'low', 'medium', 'high', 'xhigh', 'max', 'auto']] = None

Default reasoning level when supported by the model.

reasoning_budget
type: int | None = None

Default token budget reserved for reasoning.

provider
type: Literal['openrouter'] = 'openrouter'

Discriminator for OpenRouter inference settings.

RerankerConfig

Inherits: BaseModelConfig

Config for a reranker model.

Fields

id
type: str (required)

Yera-internal dot-delimited identifier. Must have at least three segments: <provider>.<author>.<model_name>, with optional further nesting for variants e.g. llama_cpp.meta.llama3.8b.q4_k_m. Does not encode the connection name — that is stored separately in connection.

model_id
type: str (required)

Provider-side model identifier passed directly to the API or library, e.g. "claude-sonnet-4-20250514" or a HuggingFace repo id like "meta-llama/Meta-Llama-3-8B-GGUF".

name
type: str (required)

Human-readable display name, e.g. "Claude Sonnet 4".

provider
type: str (required)

Key into :class:~yera.config.schema.providers.Providers identifying which provider serves this model, e.g. "aws".

connection
type: str (required)

Name of the provider connection under which this model was discovered, e.g. "work". Together with provider, uniquely identifies the runtime interface and credentials to use.

author
type: str (required)

Who created the model, e.g. "anthropic" or "meta". Distinct from provider — a Meta model can be served via AWS Bedrock.

type
type: Literal['reranker'] = 'reranker'

Discriminator field, always "reranker".

STTConfig

Inherits: BaseModelConfig

Config for a speech-to-text model.

Fields

id
type: str (required)

Yera-internal dot-delimited identifier. Must have at least three segments: <provider>.<author>.<model_name>, with optional further nesting for variants e.g. llama_cpp.meta.llama3.8b.q4_k_m. Does not encode the connection name — that is stored separately in connection.

model_id
type: str (required)

Provider-side model identifier passed directly to the API or library, e.g. "claude-sonnet-4-20250514" or a HuggingFace repo id like "meta-llama/Meta-Llama-3-8B-GGUF".

name
type: str (required)

Human-readable display name, e.g. "Claude Sonnet 4".

provider
type: str (required)

Key into :class:~yera.config.schema.providers.Providers identifying which provider serves this model, e.g. "aws".

connection
type: str (required)

Name of the provider connection under which this model was discovered, e.g. "work". Together with provider, uniquely identifies the runtime interface and credentials to use.

author
type: str (required)

Who created the model, e.g. "anthropic" or "meta". Distinct from provider — a Meta model can be served via AWS Bedrock.

type
type: Literal['stt'] = 'stt'

Discriminator field, always "stt".

TTSConfig

Inherits: BaseModelConfig

Config for a text-to-speech model.

Fields

id
type: str (required)

Yera-internal dot-delimited identifier. Must have at least three segments: <provider>.<author>.<model_name>, with optional further nesting for variants e.g. llama_cpp.meta.llama3.8b.q4_k_m. Does not encode the connection name — that is stored separately in connection.

model_id
type: str (required)

Provider-side model identifier passed directly to the API or library, e.g. "claude-sonnet-4-20250514" or a HuggingFace repo id like "meta-llama/Meta-Llama-3-8B-GGUF".

name
type: str (required)

Human-readable display name, e.g. "Claude Sonnet 4".

provider
type: str (required)

Key into :class:~yera.config.schema.providers.Providers identifying which provider serves this model, e.g. "aws".

connection
type: str (required)

Name of the provider connection under which this model was discovered, e.g. "work". Together with provider, uniquely identifies the runtime interface and credentials to use.

author
type: str (required)

Who created the model, e.g. "anthropic" or "meta". Distinct from provider — a Meta model can be served via AWS Bedrock.

type
type: Literal['tts'] = 'tts'

Discriminator field, always "tts".

TypedCatalogue

Inherits: BaseModel, Generic[M]

A filtered runtime view of models of one type for the active Yera profile.

Never serialized — constructed at runtime by :meth:ModelsUniverse.for_profile. The default field is populated from the active :class:~yera.config.schema.profiles.Profile's model_defaults.

Fields

models
type: dict[str, ~M] = <factory>

Models of this type visible under the active Yera profile.

default
type: str | None = None

Yera id of the default model for this type, sourced from the active Yera profile. None if no default is configured — a runtime error when a default is actually needed.

Methods

default_exists — Check that the default model id exists in models.

TypedCatalogue.default_exists

default_exists() → TypedCatalogue

Check that the default model id exists in models.

Returns

type: TypedCatalogue

The validated catalogue.

Raises

ValueError

If default is set but not present in models.