api.md

Shared Types

from llama_stack_client.types import (
    AgentConfig,
    CompletionMessage,
    Document,
    InterleavedContent,
    InterleavedContentItem,
    Message,
    ParamType,
    QueryConfig,
    QueryResult,
    ResponseFormat,
    SafetyViolation,
    SamplingParams,
    ScoringResult,
    SystemMessage,
    ToolCall,
    ToolResponseMessage,
    UserMessage,
)

Toolgroups

Types:

from llama_stack_client.types import ListToolGroupsResponse, ToolGroup, ToolgroupListResponse

Methods:

• client.toolgroups.list() -> ToolgroupListResponse
• client.toolgroups.get(toolgroup_id) -> ToolGroup
• client.toolgroups.register(**params) -> None
• client.toolgroups.unregister(toolgroup_id) -> None

Tools

Types:

from llama_stack_client.types import ToolListResponse

Methods:

• client.tools.list(**params) -> ToolListResponse
• client.tools.get(tool_name) -> ToolDef

ToolRuntime

Types:

from llama_stack_client.types import ToolDef, ToolInvocationResult, ToolRuntimeListToolsResponse

Methods:

• client.tool_runtime.invoke_tool(**params) -> ToolInvocationResult
• client.tool_runtime.list_tools(**params) -> ToolRuntimeListToolsResponse

RagTool

Methods:

• client.tool_runtime.rag_tool.insert(**params) -> None
• client.tool_runtime.rag_tool.query(**params) -> QueryResult

Responses

Types:

from llama_stack_client.types import (
    ResponseObject,
    ResponseObjectStream,
    ResponseListResponse,
    ResponseDeleteResponse,
)

Methods:

• client.responses.create(**params) -> ResponseObject
• client.responses.retrieve(response_id) -> ResponseObject
• client.responses.list(**params) -> SyncOpenAICursorPage[ResponseListResponse]
• client.responses.delete(response_id) -> ResponseDeleteResponse

InputItems

Types:

from llama_stack_client.types.responses import InputItemListResponse

Methods:

• client.responses.input_items.list(response_id, **params) -> InputItemListResponse

Conversations

Types:

from llama_stack_client.types import ConversationObject, ConversationDeleteResponse

Methods:

• client.conversations.create(**params) -> ConversationObject
• client.conversations.retrieve(conversation_id) -> ConversationObject
• client.conversations.update(conversation_id, **params) -> ConversationObject
• client.conversations.delete(conversation_id) -> ConversationDeleteResponse

Items

Types:

from llama_stack_client.types.conversations import (
    ItemCreateResponse,
    ItemListResponse,
    ItemGetResponse,
)

Methods:

• client.conversations.items.create(conversation_id, **params) -> ItemCreateResponse
• client.conversations.items.list(conversation_id, **params) -> ItemListResponse
• client.conversations.items.get(item_id, *, conversation_id) -> ItemGetResponse

Datasets

Types:

from llama_stack_client.types import (
    ListDatasetsResponse,
    DatasetRetrieveResponse,
    DatasetListResponse,
    DatasetIterrowsResponse,
    DatasetRegisterResponse,
)

Methods:

• client.datasets.retrieve(dataset_id) -> DatasetRetrieveResponse
• client.datasets.list() -> DatasetListResponse
• client.datasets.appendrows(dataset_id, **params) -> None
• client.datasets.iterrows(dataset_id, **params) -> DatasetIterrowsResponse
• client.datasets.register(**params) -> DatasetRegisterResponse
• client.datasets.unregister(dataset_id) -> None

Inspect

Types:

from llama_stack_client.types import HealthInfo, ProviderInfo, RouteInfo, VersionInfo

Methods:

• client.inspect.health() -> HealthInfo
• client.inspect.version() -> VersionInfo

Embeddings

Types:

from llama_stack_client.types import CreateEmbeddingsResponse

Methods:

• client.embeddings.create(**params) -> CreateEmbeddingsResponse

Chat

Types:

from llama_stack_client.types import ChatCompletionChunk

Completions

Types:

from llama_stack_client.types.chat import (
    CompletionCreateResponse,
    CompletionRetrieveResponse,
    CompletionListResponse,
)

Methods:

• client.chat.completions.create(**params) -> CompletionCreateResponse
• client.chat.completions.retrieve(completion_id) -> CompletionRetrieveResponse
• client.chat.completions.list(**params) -> SyncOpenAICursorPage[CompletionListResponse]

Completions

Types:

from llama_stack_client.types import CompletionCreateResponse

Methods:

• client.completions.create(**params) -> CompletionCreateResponse

VectorIo

Types:

from llama_stack_client.types import QueryChunksResponse

Methods:

• client.vector_io.insert(**params) -> None
• client.vector_io.query(**params) -> QueryChunksResponse

VectorStores

Types:

from llama_stack_client.types import (
    ListVectorStoresResponse,
    VectorStore,
    VectorStoreDeleteResponse,
    VectorStoreSearchResponse,
)

Methods:

• client.vector_stores.create(**params) -> VectorStore
• client.vector_stores.retrieve(vector_store_id) -> VectorStore
• client.vector_stores.update(vector_store_id, **params) -> VectorStore
• client.vector_stores.list(**params) -> SyncOpenAICursorPage[VectorStore]
• client.vector_stores.delete(vector_store_id) -> VectorStoreDeleteResponse
• client.vector_stores.search(vector_store_id, **params) -> VectorStoreSearchResponse

Files

Types:

from llama_stack_client.types.vector_stores import (
    VectorStoreFile,
    FileDeleteResponse,
    FileContentResponse,
)

Methods:

• client.vector_stores.files.create(vector_store_id, **params) -> VectorStoreFile
• client.vector_stores.files.retrieve(file_id, *, vector_store_id) -> VectorStoreFile
• client.vector_stores.files.update(file_id, *, vector_store_id, **params) -> VectorStoreFile
• client.vector_stores.files.list(vector_store_id, **params) -> SyncOpenAICursorPage[VectorStoreFile]
• client.vector_stores.files.delete(file_id, *, vector_store_id) -> FileDeleteResponse
• client.vector_stores.files.content(file_id, *, vector_store_id) -> FileContentResponse

FileBatches

Types:

from llama_stack_client.types.vector_stores import (
    ListVectorStoreFilesInBatchResponse,
    VectorStoreFileBatches,
)

Methods:

• client.vector_stores.file_batches.create(vector_store_id, **params) -> VectorStoreFileBatches
• client.vector_stores.file_batches.retrieve(batch_id, *, vector_store_id) -> VectorStoreFileBatches
• client.vector_stores.file_batches.cancel(batch_id, *, vector_store_id) -> VectorStoreFileBatches
• client.vector_stores.file_batches.list_files(batch_id, *, vector_store_id, **params) -> SyncOpenAICursorPage[VectorStoreFile]

Models

Types:

from llama_stack_client.types import ListModelsResponse, Model, ModelListResponse

Methods:

• client.models.retrieve(model_id) -> Model
• client.models.list() -> ModelListResponse
• client.models.register(**params) -> Model
• client.models.unregister(model_id) -> None

OpenAI

Methods:

• client.models.openai.list() -> ModelListResponse

Providers

Types:

from llama_stack_client.types import ListProvidersResponse, ProviderListResponse

Methods:

• client.providers.retrieve(provider_id) -> ProviderInfo
• client.providers.list() -> ProviderListResponse

Routes

Types:

from llama_stack_client.types import ListRoutesResponse, RouteListResponse

Methods:

• client.routes.list() -> RouteListResponse

Moderations

Types:

from llama_stack_client.types import CreateResponse

Methods:

• client.moderations.create(**params) -> CreateResponse

Safety

Types:

from llama_stack_client.types import RunShieldResponse

Methods:

• client.safety.run_shield(**params) -> RunShieldResponse

Shields

Types:

from llama_stack_client.types import ListShieldsResponse, Shield, ShieldListResponse

Methods:

• client.shields.retrieve(identifier) -> Shield
• client.shields.list() -> ShieldListResponse
• client.shields.delete(identifier) -> None
• client.shields.register(**params) -> Shield

SyntheticDataGeneration

Types:

from llama_stack_client.types import SyntheticDataGenerationResponse

Methods:

• client.synthetic_data_generation.generate(**params) -> SyntheticDataGenerationResponse

Telemetry

Types:

from llama_stack_client.types import (
    Event,
    QueryCondition,
    QuerySpansResponse,
    SpanWithStatus,
    Trace,
    TelemetryGetSpanResponse,
    TelemetryGetSpanTreeResponse,
    TelemetryQueryMetricsResponse,
    TelemetryQuerySpansResponse,
    TelemetryQueryTracesResponse,
)

Methods:

• client.telemetry.get_span(span_id, *, trace_id) -> TelemetryGetSpanResponse
• client.telemetry.get_span_tree(span_id, **params) -> TelemetryGetSpanTreeResponse
• client.telemetry.get_trace(trace_id) -> Trace
• client.telemetry.query_metrics(metric_name, **params) -> TelemetryQueryMetricsResponse
• client.telemetry.query_spans(**params) -> TelemetryQuerySpansResponse
• client.telemetry.query_traces(**params) -> TelemetryQueryTracesResponse
• client.telemetry.save_spans_to_dataset(**params) -> None

Scoring

Types:

from llama_stack_client.types import ScoringScoreResponse, ScoringScoreBatchResponse

Methods:

• client.scoring.score(**params) -> ScoringScoreResponse
• client.scoring.score_batch(**params) -> ScoringScoreBatchResponse

ScoringFunctions

Types:

from llama_stack_client.types import (
    ListScoringFunctionsResponse,
    ScoringFn,
    ScoringFnParams,
    ScoringFunctionListResponse,
)

Methods:

• client.scoring_functions.retrieve(scoring_fn_id) -> ScoringFn
• client.scoring_functions.list() -> ScoringFunctionListResponse
• client.scoring_functions.register(**params) -> None

Benchmarks

Types:

from llama_stack_client.types import Benchmark, ListBenchmarksResponse, BenchmarkListResponse

Methods:

• client.benchmarks.retrieve(benchmark_id) -> Benchmark
• client.benchmarks.list() -> BenchmarkListResponse
• client.benchmarks.register(**params) -> None

Files

Types:

from llama_stack_client.types import DeleteFileResponse, File, ListFilesResponse

Methods:

• client.files.create(**params) -> File
• client.files.retrieve(file_id) -> File
• client.files.list(**params) -> SyncOpenAICursorPage[File]
• client.files.delete(file_id) -> DeleteFileResponse
• client.files.content(file_id) -> object

Alpha

Inference

Types:

from llama_stack_client.types.alpha import InferenceRerankResponse

Methods:

• client.alpha.inference.rerank(**params) -> InferenceRerankResponse

PostTraining

Types:

from llama_stack_client.types.alpha import (
    AlgorithmConfig,
    ListPostTrainingJobsResponse,
    PostTrainingJob,
)

Methods:

• client.alpha.post_training.preference_optimize(**params) -> PostTrainingJob
• client.alpha.post_training.supervised_fine_tune(**params) -> PostTrainingJob

Job

Types:

from llama_stack_client.types.alpha.post_training import (
    JobListResponse,
    JobArtifactsResponse,
    JobStatusResponse,
)

Methods:

• client.alpha.post_training.job.list() -> JobListResponse
• client.alpha.post_training.job.artifacts(**params) -> JobArtifactsResponse
• client.alpha.post_training.job.cancel(**params) -> None
• client.alpha.post_training.job.status(**params) -> JobStatusResponse

Eval

Types:

from llama_stack_client.types.alpha import BenchmarkConfig, EvaluateResponse, Job

Methods:

• client.alpha.eval.evaluate_rows(benchmark_id, **params) -> EvaluateResponse
• client.alpha.eval.evaluate_rows_alpha(benchmark_id, **params) -> EvaluateResponse
• client.alpha.eval.run_eval(benchmark_id, **params) -> Job
• client.alpha.eval.run_eval_alpha(benchmark_id, **params) -> Job

Jobs

Methods:

• client.alpha.eval.jobs.retrieve(job_id, *, benchmark_id) -> EvaluateResponse
• client.alpha.eval.jobs.cancel(job_id, *, benchmark_id) -> None
• client.alpha.eval.jobs.status(job_id, *, benchmark_id) -> Job

Agents

Types:

from llama_stack_client.types.alpha import (
    InferenceStep,
    MemoryRetrievalStep,
    ShieldCallStep,
    ToolExecutionStep,
    ToolResponse,
    AgentCreateResponse,
    AgentRetrieveResponse,
    AgentListResponse,
)

Methods:

• client.alpha.agents.create(**params) -> AgentCreateResponse
• client.alpha.agents.retrieve(agent_id) -> AgentRetrieveResponse
• client.alpha.agents.list(**params) -> AgentListResponse
• client.alpha.agents.delete(agent_id) -> None

Session

Types:

from llama_stack_client.types.alpha.agents import (
    Session,
    SessionCreateResponse,
    SessionListResponse,
)

Methods:

• client.alpha.agents.session.create(agent_id, **params) -> SessionCreateResponse
• client.alpha.agents.session.retrieve(session_id, *, agent_id, **params) -> Session
• client.alpha.agents.session.list(agent_id, **params) -> SessionListResponse
• client.alpha.agents.session.delete(session_id, *, agent_id) -> None

Steps

Types:

from llama_stack_client.types.alpha.agents import StepRetrieveResponse

Methods:

• client.alpha.agents.steps.retrieve(step_id, *, agent_id, session_id, turn_id) -> StepRetrieveResponse

Turn

Types:

from llama_stack_client.types.alpha.agents import (
    AgentTurnResponseStreamChunk,
    Turn,
    TurnResponseEvent,
)

Methods:

• client.alpha.agents.turn.create(session_id, *, agent_id, **params) -> Turn
• client.alpha.agents.turn.retrieve(turn_id, *, agent_id, session_id) -> Turn
• client.alpha.agents.turn.resume(turn_id, *, agent_id, session_id, **params) -> Turn