orca-sdk 0.1.2__py3-none-any.whl → 0.1.4__py3-none-any.whl

orca_sdk/async_client.py

@@ -0,0 +1,3942 @@
+# Do not edit this file manually! It is generated by scripts/codegen.py
+
+from __future__ import annotations
+
+import json
+import logging
+import os
+import uuid
+from contextlib import contextmanager
+from contextvars import ContextVar
+from string import Formatter
+from typing import (
+    Any,
+    Callable,
+    Generator,
+    Literal,
+    Mapping,
+    NotRequired,
+    Self,
+    TypedDict,
+    cast,
+    overload,
+)
+
+from httpx import (
+    URL,
+    USE_CLIENT_DEFAULT,
+    AsyncBaseTransport,
+    AsyncClient,
+    AsyncHTTPTransport,
+    BaseTransport,
+    Headers,
+    HTTPTransport,
+    Limits,
+    Proxy,
+    Request,
+    Response,
+    Timeout,
+)
+from httpx._client import UseClientDefault  # type: ignore
+from httpx._types import AuthTypes  # type: ignore
+from httpx._types import CookieTypes  # type: ignore
+from httpx._types import FileTypes  # type: ignore
+from httpx._types import HeaderTypes  # type: ignore
+from httpx._types import RequestContent  # type: ignore
+from httpx._types import RequestExtensions  # type: ignore
+from httpx._types import TimeoutTypes  # type: ignore
+from httpx_retries import Retry, RetryTransport
+
+
+class ActionRecommendation(TypedDict):
+    action: Literal["remove_duplicates", "detect_mislabels", "add_memories", "finetuning"]
+    """
+    The recommended action to take
+    """
+    rationale: str
+    """
+    Explanation for why this action was recommended
+    """
+
+
+class AddMemorySuggestion(TypedDict):
+    value: str
+    label_name: str
+
+
+class AliveResponse(TypedDict):
+    ok: bool
+
+
+class ApiKeyMetadata(TypedDict):
+    id: str
+    org_id: str
+    name: str
+    created_by: str | None
+    created_at: str
+    updated_at: str
+    scope: list[Literal["ADMINISTER", "PREDICT"]]
+
+
+class BaseLabelPredictionResult(TypedDict):
+    prediction_id: str | None
+    confidence: float
+    anomaly_score: float | None
+    label: int | None
+    label_name: str | None
+    logits: list[float]
+
+
+class BaseModel(TypedDict):
+    pass
+
+
+class BaseScorePredictionResult(TypedDict):
+    prediction_id: str | None
+    confidence: float
+    anomaly_score: float | None
+    score: float | None
+
+
+class CascadeEditSuggestionsRequest(TypedDict):
+    old_label: int | None
+    new_label: int
+    max_neighbors: NotRequired[int]
+    max_validation_neighbors: NotRequired[int]
+    similarity_threshold: NotRequired[float | None]
+    only_if_has_old_label: NotRequired[bool]
+    exclude_if_new_label: NotRequired[bool]
+    suggestion_cooldown_time: NotRequired[float]
+    label_confirmation_cooldown_time: NotRequired[float]
+
+
+class ClassPattern(TypedDict):
+    label: int
+    name: str
+    description: str
+
+
+class ClassPatternsDescription(TypedDict):
+    overview: str
+    classes: list[ClassPattern]
+    summary: str
+
+
+class ClassRepresentatives(TypedDict):
+    label: int
+    label_name: str | None
+    representative_memory_ids: list[str]
+
+
+class ClassificationEvaluationRequest(TypedDict):
+    datasource_name_or_id: str
+    memoryset_override_name_or_id: NotRequired[str | None]
+    datasource_label_column: str
+    datasource_value_column: str
+    record_telemetry: NotRequired[bool]
+    telemetry_tags: NotRequired[list[str] | None]
+    subsample: NotRequired[int | float | None]
+    ignore_unlabeled: NotRequired[bool]
+
+
+class CleanupResponse(TypedDict):
+    deleted_milvus_collections: list[str]
+    deleted_milvus_collections_count: int
+    deleted_blob_paths: list[str]
+    deleted_blob_paths_count: int
+
+
+class ClusterMetrics(TypedDict):
+    cluster: int
+    memory_count: int
+
+
+ColumnType = Literal["STRING", "FLOAT", "INT", "BOOL", "ENUM", "IMAGE", "OTHER"]
+
+
+class ConstraintViolationErrorResponse(TypedDict):
+    status_code: NotRequired[int]
+    constraint: str
+
+
+class CountPredictionsRequest(TypedDict):
+    model_id: NotRequired[str | None]
+    tag: NotRequired[str | None]
+    prediction_ids: NotRequired[list[str] | None]
+    start_timestamp: NotRequired[str | None]
+    end_timestamp: NotRequired[str | None]
+    memory_id: NotRequired[str | None]
+
+
+class CreateApiKeyRequest(TypedDict):
+    id: NotRequired[str]
+    name: NotRequired[str]
+    created_by: NotRequired[str | None]
+    scope: list[Literal["ADMINISTER", "PREDICT"]]
+
+
+class CreateApiKeyResponse(TypedDict):
+    id: str
+    org_id: str
+    name: str
+    created_by: str | None
+    created_at: str
+    updated_at: str
+    scope: list[Literal["ADMINISTER", "PREDICT"]]
+    api_key: str
+
+
+class CreateDatasourceFromContentRequest(TypedDict):
+    name: str
+    description: NotRequired[str | None]
+    content: Any
+
+
+class CreateOrgPlanRequest(TypedDict):
+    tier: Literal["FREE", "PRO", "ENTERPRISE", "CANCELLED"]
+
+
+class DatasetFilterItem(TypedDict):
+    field: str
+    op: Literal["==", "!=", ">", ">=", "<", "<=", "in", "not in", "like"]
+    value: Any
+
+
+class DeleteMemoriesRequest(TypedDict):
+    memory_ids: list[str]
+
+
+class DeleteMemorysetsRequest(TypedDict):
+    memoryset_ids: list[str]
+
+
+class EmbedRequest(TypedDict):
+    values: list[str] | list[bytes] | list[str | bytes]
+    max_seq_length: NotRequired[int | None]
+    prompt: NotRequired[str | None]
+
+
+class EmbeddingEvaluationRequest(TypedDict):
+    datasource_name_or_id: str
+    eval_datasource_name_or_id: NotRequired[str | None]
+    subsample: NotRequired[int | float | None]
+    datasource_value_column: NotRequired[str]
+    datasource_label_column: NotRequired[str | None]
+    datasource_score_column: NotRequired[str | None]
+    neighbor_count: NotRequired[int]
+    batch_size: NotRequired[int]
+    weigh_memories: NotRequired[bool]
+
+
+EmbeddingFinetuningMethod = Literal["classification", "regression", "batch_triplet_loss"]
+
+
+class FeedbackMetrics(TypedDict):
+    avg: float | None
+    count: int
+
+
+FeedbackType = Literal["CONTINUOUS", "BINARY"]
+
+
+class FilterItem(TypedDict):
+    field: list
+    op: Literal["==", "!=", ">", ">=", "<", "<=", "in", "not in", "like"]
+    value: str | int | float | bool | list[str | None] | list[int] | list[float] | list[bool] | None
+
+
+class GetDatasourceRowCountRequest(TypedDict):
+    filters: NotRequired[list[DatasetFilterItem]]
+
+
+class GetDatasourceRowsRequest(TypedDict):
+    filters: NotRequired[list[DatasetFilterItem]]
+    limit: NotRequired[int]
+    offset: NotRequired[int]
+    shuffle: NotRequired[bool]
+    shuffle_seed: NotRequired[int | None]
+
+
+class GetMemoriesRequest(TypedDict):
+    memory_ids: list[str]
+
+
+class HealthyResponse(TypedDict):
+    ok: bool
+    checks: dict[str, bool]
+    durations: dict[str, int]
+    draining: bool
+    config: dict[str, str | float | int | bool | None]
+
+
+class InternalServerErrorResponse(TypedDict):
+    status_code: NotRequired[int]
+    message: str
+    request_id: str
+
+
+JobStatus = Literal["INITIALIZED", "DISPATCHED", "WAITING", "PROCESSING", "COMPLETED", "FAILED", "ABORTING", "ABORTED"]
+
+
+class JobStatusInfo(TypedDict):
+    status: JobStatus
+    steps_total: int | None
+    steps_completed: int | None
+    exception: str | None
+    updated_at: str
+    created_at: str
+
+
+class LabelClassMetrics(TypedDict):
+    label: int | None
+    label_name: NotRequired[str | None]
+    average_lookup_score: float
+    memory_count: int
+
+
+class LabelPercentage(TypedDict):
+    label: int | None
+    percentage: float
+
+
+class LabeledExample(TypedDict):
+    text: str
+    label_name: str
+
+
+class LabeledMemoryInsert(TypedDict):
+    memory_id: NotRequired[str | None]
+    value: str | bytes
+    metadata: NotRequired[dict[str, str | int | float | bool | None]]
+    source_id: NotRequired[str | None]
+    partition_id: NotRequired[str | None]
+    label: int | None
+
+
+class ListMemoriesRequest(TypedDict):
+    offset: NotRequired[int]
+    limit: NotRequired[int]
+    filters: NotRequired[list[FilterItem]]
+
+
+class LookupRequest(TypedDict):
+    query: list[str]
+    count: NotRequired[int]
+    prompt: NotRequired[str | None]
+
+
+class LookupScoreMetrics(TypedDict):
+    median: float
+    std: float
+    quantiles: list[float]
+    quantile_values: list[float]
+
+
+class MemoryMetrics(TypedDict):
+    is_duplicate: NotRequired[bool]
+    duplicate_memory_ids: NotRequired[list[str]]
+    has_potential_duplicates: NotRequired[bool]
+    potential_duplicate_memory_ids: NotRequired[list[str] | None]
+    cluster: NotRequired[int]
+    embedding_2d: NotRequired[list]
+    anomaly_score: NotRequired[float]
+    neighbor_label_logits: NotRequired[list[float]]
+    neighbor_predicted_label: NotRequired[int | None]
+    neighbor_predicted_label_ambiguity: NotRequired[float]
+    neighbor_predicted_label_confidence: NotRequired[float]
+    current_label_neighbor_confidence: NotRequired[float]
+    normalized_neighbor_label_entropy: NotRequired[float]
+    neighbor_predicted_label_matches_current_label: NotRequired[bool | None]
+    spread: NotRequired[float]
+    uniformity: NotRequired[float]
+    concept_id: NotRequired[int | None]
+    subconcept_id: NotRequired[int | None]
+
+
+MemoryType = Literal["LABELED", "SCORED"]
+
+
+class MemorysetClassPatternsAnalysisConfig(TypedDict):
+    representatives_per_class: NotRequired[int]
+    enable_patterns_description: NotRequired[bool]
+
+
+class MemorysetClassPatternsMetrics(TypedDict):
+    class_representatives: list[ClassRepresentatives]
+    patterns_description: NotRequired[ClassPatternsDescription | None]
+    mean_spread: float
+    variance_spread: float
+    mean_uniformity: float
+    variance_uniformity: float
+    updated_at: str
+
+
+class MemorysetClusterAnalysisConfig(TypedDict):
+    min_cluster_size: NotRequired[int | None]
+    max_cluster_size: NotRequired[int | None]
+    partitioning_method: NotRequired[Literal["ng", "rb", "cpm"]]
+    resolution: NotRequired[float | None]
+    num_iterations: NotRequired[int]
+    random_state: NotRequired[int | None]
+
+
+class MemorysetClusterMetrics(TypedDict):
+    cluster_metrics: list[ClusterMetrics]
+    num_outliers: int
+    num_clusters: int
+    updated_at: str
+
+
+class MemorysetConceptAnalysisConfig(TypedDict):
+    high_level_description: NotRequired[str | None]
+    max_sample_rows: NotRequired[int]
+    min_lookup_count: NotRequired[int]
+    max_lookup_count: NotRequired[int]
+    overlap_decay_tau: NotRequired[float]
+    target_lcc_fraction: NotRequired[float]
+    min_cluster_count: NotRequired[int]
+    max_cluster_count: NotRequired[int]
+    max_trial_count: NotRequired[int]
+    resolution: NotRequired[float]
+    iterations: NotRequired[int]
+    use_generative_naming: NotRequired[bool]
+    naming_examples_count: NotRequired[int]
+    naming_counterexample_count: NotRequired[int]
+    primary_label_pct_threshold: NotRequired[float]
+    seed: NotRequired[int]
+
+
+class MemorysetDistributionAnalysisConfig(TypedDict):
+    neighbor_counts: NotRequired[list[int]]
+    quantiles: NotRequired[list[float]]
+
+
+class MemorysetDistributionMetrics(TypedDict):
+    lookup_score_metrics: dict[str, LookupScoreMetrics]
+    updated_at: str
+
+
+class MemorysetDuplicateAnalysisConfig(TypedDict):
+    potential_duplicate_threshold: NotRequired[float]
+
+
+class MemorysetDuplicateMetrics(TypedDict):
+    num_duplicates: int
+    num_potential_duplicates: int
+    updated_at: str
+
+
+class MemorysetLabelAnalysisConfig(TypedDict):
+    normalize_logits: NotRequired[bool]
+
+
+class MemorysetLabelMetrics(TypedDict):
+    label_metrics: list[LabelClassMetrics]
+    neighbor_prediction_accuracy: float
+    mean_neighbor_label_confidence: float
+    mean_neighbor_label_entropy: float
+    mean_neighbor_predicted_label_ambiguity: float
+    num_no_labeled_neighbors: int
+    num_potential_mislabels: int
+    updated_at: str
+
+
+class MemorysetProjectionAnalysisConfig(TypedDict):
+    min_dist: NotRequired[float]
+    spread: NotRequired[float]
+
+
+class MemorysetProjectionMetrics(TypedDict):
+    updated_at: str
+
+
+class MemorysetUpdate(TypedDict):
+    label_names: NotRequired[list[str]]
+    description: NotRequired[str | None]
+    name: NotRequired[str]
+    notes: NotRequired[str | None]
+    hidden: NotRequired[bool]
+
+
+class NotFoundErrorResponse(TypedDict):
+    status_code: NotRequired[int]
+    resource: (
+        Literal[
+            "org",
+            "api_key",
+            "datasource",
+            "memoryset",
+            "classification_model",
+            "regression_model",
+            "prediction",
+            "memory",
+            "evaluation",
+            "analysis",
+            "job",
+            "pretrained_embedding_model",
+            "finetuned_embedding_model",
+            "feedback_category",
+            "org_plan",
+            "worker",
+        ]
+        | None
+    )
+
+
+class OrgPlan(TypedDict):
+    org_id: str
+    created_at: str
+    updated_at: str
+    tier: Literal["FREE", "PRO", "ENTERPRISE", "CANCELLED"]
+
+
+class PRCurve(TypedDict):
+    thresholds: list[float]
+    precisions: list[float]
+    recalls: list[float]
+
+
+class PredictionFeedback(TypedDict):
+    prediction_id: str
+    category_name: str
+    value: float | bool
+    comment: str | None
+    id: str
+    org_id: str
+    category_id: str
+    category_type: FeedbackType
+    created_at: str
+    updated_at: str
+
+
+class PredictionFeedbackCategory(TypedDict):
+    id: str
+    org_id: str
+    name: str
+    type: FeedbackType
+    created_at: str
+    updated_at: str
+
+
+class PredictionFeedbackRequest(TypedDict):
+    prediction_id: str
+    category_name: str
+    value: NotRequired[float | bool | None]
+    """
+    The feedback value. For updates, UNSET means keep existing value. None means delete the feedback.
+    """
+    comment: NotRequired[str | None]
+    """
+    Optional comment. For updates, UNSET means keep existing comment. None means remove the comment.
+    """
+
+
+class PredictionFeedbackResult(TypedDict):
+    prediction_ids: list[str]
+    deleted_feedback_ids: list[str]
+    updated_feedback_ids: list[str]
+    inserted_feedback_ids: list[str]
+    new_category_ids: list[str]
+
+
+PredictionSort = list[list]
+
+
+class PredictiveModelUpdate(TypedDict):
+    description: NotRequired[str | None]
+    notes: NotRequired[str | None]
+    locked: NotRequired[bool]
+
+
+PretrainedEmbeddingModelName = Literal[
+    "CLIP_BASE",
+    "GTE_BASE",
+    "CDE_SMALL",
+    "DISTILBERT",
+    "GTE_SMALL",
+    "MXBAI_LARGE",
+    "E5_LARGE",
+    "QWEN2_1_5B",
+    "BGE_BASE",
+    "GIST_LARGE",
+]
+
+
+RACHeadType = Literal["KNN", "MMOE", "FF", "BMMOE"]
+
+
+RARHeadType = Literal["MMOE", "KNN"]
+
+
+class ROCCurve(TypedDict):
+    thresholds: list[float]
+    false_positive_rates: list[float]
+    true_positive_rates: list[float]
+
+
+class ReadyResponse(TypedDict):
+    ok: bool
+    draining: bool
+
+
+class RegressionEvaluationRequest(TypedDict):
+    datasource_name_or_id: str
+    memoryset_override_name_or_id: NotRequired[str | None]
+    datasource_score_column: str
+    datasource_value_column: str
+    record_telemetry: NotRequired[bool]
+    telemetry_tags: NotRequired[list[str] | None]
+    subsample: NotRequired[int | float | None]
+    ignore_unlabeled: NotRequired[bool]
+
+
+class RegressionMetrics(TypedDict):
+    coverage: float
+    mse: float
+    rmse: float
+    mae: float
+    r2: float
+    explained_variance: float
+    loss: float
+    anomaly_score_mean: NotRequired[float | None]
+    anomaly_score_median: NotRequired[float | None]
+    anomaly_score_variance: NotRequired[float | None]
+
+
+class RegressionModelMetadata(TypedDict):
+    id: str
+    org_id: str
+    name: str
+    description: str | None
+    notes: str | None
+    version: int
+    memoryset_id: str
+    memory_lookup_count: int
+    storage_path: str
+    memoryset_collection_name: str
+    created_at: str
+    updated_at: str
+    locked: bool
+    head_type: RARHeadType
+
+
+class RegressionPredictionRequest(TypedDict):
+    input_values: list[str] | list[bytes] | list[str | bytes]
+    expected_scores: NotRequired[list[float] | None]
+    tags: NotRequired[list[str]]
+    memoryset_override_name_or_id: NotRequired[str | None]
+    save_telemetry: NotRequired[bool]
+    save_telemetry_synchronously: NotRequired[bool]
+    prompt: NotRequired[str | None]
+    use_lookup_cache: NotRequired[bool]
+    consistency_level: NotRequired[Literal["Bounded", "Session", "Strong", "Eventual"] | None]
+    ignore_unlabeled: NotRequired[bool]
+
+
+class ScorePredictionMemoryLookup(TypedDict):
+    value: str | bytes
+    embedding: list[float]
+    source_id: str | None
+    partition_id: str | None
+    metadata: dict[str, str | int | float | bool | None]
+    memory_id: str
+    memory_version: int
+    created_at: str
+    updated_at: str
+    edited_at: str
+    metrics: MemoryMetrics
+    lookup_score: float
+    score: float | None
+    prediction_id: str
+    attention_weight: float
+
+
+class ScorePredictionWithMemoriesAndFeedback(TypedDict):
+    prediction_id: str
+    confidence: float
+    anomaly_score: float | None
+    score: float | None
+    timestamp: str
+    input_value: str | bytes
+    input_embedding: list[float]
+    expected_score: float | None
+    memories: list[ScorePredictionMemoryLookup]
+    org_id: str
+    memoryset_id: str
+    model_id: str
+    updated_at: str
+    tags: list[str]
+    explanation: str | None
+    memory_id: str | None
+    feedbacks: list[PredictionFeedback]
+
+
+class ScoredMemory(TypedDict):
+    value: str | bytes
+    embedding: list[float]
+    source_id: str | None
+    partition_id: str | None
+    metadata: dict[str, str | int | float | bool | None]
+    memory_id: str
+    memory_version: int
+    created_at: str
+    updated_at: str
+    edited_at: str
+    metrics: MemoryMetrics
+    score: float | None
+
+
+class ScoredMemoryInsert(TypedDict):
+    memory_id: NotRequired[str | None]
+    value: str | bytes
+    metadata: NotRequired[dict[str, str | int | float | bool | None]]
+    source_id: NotRequired[str | None]
+    partition_id: NotRequired[str | None]
+    score: float | None
+
+
+class ScoredMemoryLookup(TypedDict):
+    value: str | bytes
+    embedding: list[float]
+    source_id: str | None
+    partition_id: str | None
+    metadata: dict[str, str | int | float | bool | None]
+    memory_id: str
+    memory_version: int
+    created_at: str
+    updated_at: str
+    edited_at: str
+    metrics: MemoryMetrics
+    lookup_score: float
+    score: float | None
+
+
+class ScoredMemoryUpdate(TypedDict):
+    memory_id: str
+    value: NotRequired[str | bytes]
+    metadata: NotRequired[dict[str, str | int | float | bool | None] | None]
+    source_id: NotRequired[str | None]
+    partition_id: NotRequired[str | None]
+    metrics: NotRequired[MemoryMetrics | None]
+    score: NotRequired[float | None]
+
+
+class ScoredMemoryWithFeedbackMetrics(TypedDict):
+    value: str | bytes
+    embedding: list[float]
+    source_id: str | None
+    partition_id: str | None
+    metadata: dict[str, str | int | float | bool | None]
+    memory_id: str
+    memory_version: int
+    created_at: str
+    updated_at: str
+    edited_at: str
+    metrics: MemoryMetrics
+    score: float | None
+    feedback_metrics: dict[str, FeedbackMetrics]
+    lookup_count: int
+
+
+class ServiceUnavailableErrorResponse(TypedDict):
+    status_code: NotRequired[int]
+    service: str
+
+
+class SubConceptMetrics(TypedDict):
+    id: int
+    name: str
+    description: str | None
+    primary_label: int | None
+    memory_count: int
+
+
+TelemetryField = list
+
+
+class TelemetryFilterItem(TypedDict):
+    field: TelemetryField
+    op: Literal["==", "!=", ">", ">=", "<", "<=", "in", "not in"]
+    value: float | list[float] | int | list[int]
+
+
+class TelemetrySortOptions(TypedDict):
+    field: TelemetryField
+    direction: Literal["asc", "desc"]
+
+
+class UnauthenticatedErrorResponse(TypedDict):
+    status_code: NotRequired[int]
+
+
+class UnauthorizedErrorResponse(TypedDict):
+    status_code: NotRequired[int]
+    reason: str
+
+
+class UpdateOrgPlanRequest(TypedDict):
+    tier: Literal["FREE", "PRO", "ENTERPRISE", "CANCELLED"]
+
+
+class UpdatePredictionRequest(TypedDict):
+    expected_label: NotRequired[int | None]
+    expected_score: NotRequired[float | None]
+    tags: NotRequired[list[str]]
+    memory_id: NotRequired[str | None]
+
+
+class ValidationError(TypedDict):
+    loc: list[str | int]
+    msg: str
+    type: str
+
+
+WorkerStatus = Literal["IDLE", "BUSY", "DRAINING", "SHUTDOWN", "CRASHED"]
+
+
+class GetTestErrorByStatusCodeParams(TypedDict):
+    status_code: int | Literal["error", "warning"]
+
+
+class DeleteAuthApiKeyByNameOrIdParams(TypedDict):
+    name_or_id: str
+
+
+class GetMemorysetParams(TypedDict):
+    type: NotRequired[MemoryType | None]
+    show_hidden: NotRequired[bool | None]
+
+
+class PostMemorysetByNameOrIdCloneParams(TypedDict):
+    name_or_id: str
+
+
+class PatchMemorysetByNameOrIdParams(TypedDict):
+    name_or_id: str
+
+
+class GetMemorysetByNameOrIdParams(TypedDict):
+    name_or_id: str
+
+
+class DeleteMemorysetByNameOrIdParams(TypedDict):
+    name_or_id: str
+
+
+class PostGpuMemorysetByNameOrIdLookupParams(TypedDict):
+    name_or_id: str
+
+
+class GetMemorysetByNameOrIdMemoryByMemoryIdParams(TypedDict):
+    name_or_id: str
+    memory_id: str
+    """
+    ID of the memory
+    """
+
+
+class DeleteMemorysetByNameOrIdMemoryByMemoryIdParams(TypedDict):
+    name_or_id: str
+    memory_id: str
+    """
+    ID of the memory
+    """
+
+
+class GetMemorysetByNameOrIdPotentialDuplicateGroupsParams(TypedDict):
+    name_or_id: str
+
+
+class PostMemorysetByNameOrIdMemoriesGetParams(TypedDict):
+    name_or_id: str
+
+
+class PostMemorysetByNameOrIdMemoriesParams(TypedDict):
+    name_or_id: str
+
+
+class PostMemorysetByNameOrIdMemoriesDeleteParams(TypedDict):
+    name_or_id: str
+
+
+class PatchGpuMemorysetByNameOrIdMemoryParams(TypedDict):
+    name_or_id: str
+
+
+class PostGpuMemorysetByNameOrIdMemoryParams(TypedDict):
+    name_or_id: str
+
+
+PostGpuMemorysetByNameOrIdMemoryRequest = list[LabeledMemoryInsert] | list[ScoredMemoryInsert]
+
+
+class PatchGpuMemorysetByNameOrIdMemoriesParams(TypedDict):
+    name_or_id: str
+
+
+class PostMemorysetByNameOrIdAnalysisParams(TypedDict):
+    name_or_id: str
+
+
+class GetMemorysetByNameOrIdAnalysisParams(TypedDict):
+    name_or_id: str
+    status: NotRequired[JobStatus | None]
+    limit: NotRequired[int | None]
+    offset: NotRequired[int | None]
+
+
+class GetMemorysetByNameOrIdAnalysisByAnalysisJobIdParams(TypedDict):
+    name_or_id: str
+    analysis_job_id: str
+
+
+class PostMemorysetByNameOrIdMemoryByMemoryIdCascadingEditsParams(TypedDict):
+    name_or_id: str
+    memory_id: str
+
+
+class GetFinetunedEmbeddingModelByNameOrIdParams(TypedDict):
+    name_or_id: str
+
+
+class DeleteFinetunedEmbeddingModelByNameOrIdParams(TypedDict):
+    name_or_id: str
+
+
+class PostGpuFinetunedEmbeddingModelByNameOrIdEmbeddingParams(TypedDict):
+    name_or_id: str
+
+
+class GetPretrainedEmbeddingModelByModelNameParams(TypedDict):
+    model_name: PretrainedEmbeddingModelName
+
+
+class PostGpuPretrainedEmbeddingModelByModelNameEmbeddingParams(TypedDict):
+    model_name: PretrainedEmbeddingModelName
+
+
+class PostFinetunedEmbeddingModelByNameOrIdEvaluationParams(TypedDict):
+    name_or_id: str
+
+
+class PostPretrainedEmbeddingModelByModelNameEvaluationParams(TypedDict):
+    model_name: PretrainedEmbeddingModelName
+
+
+class GetFinetunedEmbeddingModelByNameOrIdEvaluationByJobIdParams(TypedDict):
+    name_or_id: str
+    job_id: str
+
+
+class GetPretrainedEmbeddingModelByModelNameEvaluationByJobIdParams(TypedDict):
+    model_name: PretrainedEmbeddingModelName
+    job_id: str
+
+
+class GetFinetunedEmbeddingModelByNameOrIdEvaluationsParams(TypedDict):
+    name_or_id: str
+    datasource: NotRequired[str | None]
+    value_column: NotRequired[str | None]
+    label_column: NotRequired[str | None]
+    score_column: NotRequired[str | None]
+
+
+class GetPretrainedEmbeddingModelByModelNameEvaluationsParams(TypedDict):
+    model_name: PretrainedEmbeddingModelName
+    datasource: NotRequired[str | None]
+    value_column: NotRequired[str | None]
+    label_column: NotRequired[str | None]
+    score_column: NotRequired[str | None]
+
+
+class PostDatasourceUploadRequest(TypedDict):
+    name: str
+    """
+    Name for the datasource
+    """
+    description: NotRequired[str | None]
+    """
+    Optional description for the datasource
+    """
+
+
+class GetDatasourceByNameOrIdParams(TypedDict):
+    name_or_id: str
+
+
+class DeleteDatasourceByNameOrIdParams(TypedDict):
+    name_or_id: str
+
+
+class PostDatasourceByNameOrIdRowsParams(TypedDict):
+    name_or_id: str
+
+
+class PostDatasourceByNameOrIdRowsCountParams(TypedDict):
+    name_or_id: str
+
+
+class GetDatasourceByNameOrIdEmbeddingModelEvaluationsParams(TypedDict):
+    name_or_id: str
+    value_column: NotRequired[str | None]
+    label_column: NotRequired[str | None]
+    score_column: NotRequired[str | None]
+
+
+class GetDatasourceByNameOrIdDownloadParams(TypedDict):
+    name_or_id: str
+    file_type: NotRequired[Literal["hf_dataset", "json", "csv"]]
+    """
+    File type to download:
+    * `hf_dataset`: Zipped HuggingFace dataset (default)
+    * `json`: Row-oriented JSON array
+    * `csv`: CSV file
+    """
+
+
+class PatchClassificationModelByNameOrIdParams(TypedDict):
+    name_or_id: str
+
+
+class GetClassificationModelByNameOrIdParams(TypedDict):
+    name_or_id: str
+
+
+class DeleteClassificationModelByNameOrIdParams(TypedDict):
+    name_or_id: str
+
+
+class PatchRegressionModelByNameOrIdParams(TypedDict):
+    name_or_id: str
+
+
+class GetRegressionModelByNameOrIdParams(TypedDict):
+    name_or_id: str
+
+
+class DeleteRegressionModelByNameOrIdParams(TypedDict):
+    name_or_id: str
+
+
+class PostGpuClassificationModelByNameOrIdPredictionParams(TypedDict):
+    name_or_id: str
+
+
+class PostClassificationModelByNameOrIdPredictionParams(TypedDict):
+    name_or_id: str
+
+
+class PostGpuRegressionModelByNameOrIdPredictionParams(TypedDict):
+    name_or_id: str
+
+
+class PostRegressionModelByNameOrIdPredictionParams(TypedDict):
+    name_or_id: str
+
+
+class PostClassificationModelByModelNameOrIdEvaluationParams(TypedDict):
+    model_name_or_id: str
+
+
+class GetClassificationModelByModelNameOrIdEvaluationParams(TypedDict):
+    model_name_or_id: str
+
+
+class PostRegressionModelByModelNameOrIdEvaluationParams(TypedDict):
+    model_name_or_id: str
+
+
+class GetRegressionModelByModelNameOrIdEvaluationParams(TypedDict):
+    model_name_or_id: str
+
+
+class GetClassificationModelByModelNameOrIdEvaluationByJobIdParams(TypedDict):
+    model_name_or_id: str
+    job_id: str
+
+
+class DeleteClassificationModelByModelNameOrIdEvaluationByJobIdParams(TypedDict):
+    model_name_or_id: str
+    job_id: str
+
+
+class GetRegressionModelByModelNameOrIdEvaluationByJobIdParams(TypedDict):
+    model_name_or_id: str
+    job_id: str
+
+
+class DeleteRegressionModelByModelNameOrIdEvaluationByJobIdParams(TypedDict):
+    model_name_or_id: str
+    job_id: str
+
+
+class GetJobByJobIdParams(TypedDict):
+    job_id: str
+
+
+class GetJobByJobIdStatusParams(TypedDict):
+    job_id: str
+
+
+class GetJobParams(TypedDict):
+    status: NotRequired[JobStatus | list[JobStatus] | None]
+    type: NotRequired[str | list[str] | None]
+    limit: NotRequired[int | None]
+    offset: NotRequired[int]
+    start_timestamp: NotRequired[str | None]
+    end_timestamp: NotRequired[str | None]
+
+
+class DeleteJobByJobIdAbortParams(TypedDict):
+    job_id: str
+
+
+class GetWorkerParams(TypedDict):
+    status: NotRequired[WorkerStatus | list[WorkerStatus] | None]
+    limit: NotRequired[int | None]
+    offset: NotRequired[int]
+
+
+class GetWorkerByWorkerIdParams(TypedDict):
+    worker_id: str
+
+
+class GetTelemetryPredictionByPredictionIdParams(TypedDict):
+    prediction_id: str
+
+
+class PatchTelemetryPredictionByPredictionIdParams(TypedDict):
+    prediction_id: str
+
+
+class GetTelemetryPredictionByPredictionIdExplanationParams(TypedDict):
+    prediction_id: str
+    refresh: NotRequired[bool]
+
+
+class GetTelemetryPredictionByPredictionIdActionParams(TypedDict):
+    prediction_id: str
+
+
+class GetTelemetryPredictionByPredictionIdMemorySuggestionsParams(TypedDict):
+    prediction_id: str
+    """
+    ID of the prediction to generate suggestions for
+    """
+    num_memories: NotRequired[int]
+    """
+    Number of memory suggestions to generate
+    """
+    refresh: NotRequired[bool]
+    """
+    Force the explanation agent to re-run even if a cached explanation exists
+    """
+
+
+class GetTelemetryFeedbackCategoryByNameOrIdParams(TypedDict):
+    name_or_id: str
+
+
+class DeleteTelemetryFeedbackCategoryByNameOrIdParams(TypedDict):
+    name_or_id: str
+
+
+PutTelemetryPredictionFeedbackRequest = list[PredictionFeedbackRequest]
+
+
+class GetAgentsBootstrapClassificationModelByJobIdParams(TypedDict):
+    job_id: str
+
+
+class FieldValidationError(TypedDict):
+    loc: list[str | int]
+    msg: str
+    type: NotRequired[str]
+
+
+class AddMemoryRecommendations(TypedDict):
+    memories: list[AddMemorySuggestion]
+
+
+class BootstrapClassificationModelRequest(TypedDict):
+    model_description: str
+    label_names: list[str]
+    initial_examples: NotRequired[list[LabeledExample]]
+    num_examples_per_label: NotRequired[int]
+
+
+class BootstrapClassificationModelResult(TypedDict):
+    model_description: str
+    label_names: list[str]
+    model_name: str
+    generated_examples: NotRequired[list[LabeledExample]]
+
+
+class ClassificationMetrics(TypedDict):
+    coverage: float
+    f1_score: float
+    accuracy: float
+    loss: float | None
+    anomaly_score_mean: NotRequired[float | None]
+    anomaly_score_median: NotRequired[float | None]
+    anomaly_score_variance: NotRequired[float | None]
+    roc_auc: NotRequired[float | None]
+    pr_auc: NotRequired[float | None]
+    pr_curve: NotRequired[PRCurve | None]
+    roc_curve: NotRequired[ROCCurve | None]
+
+
+class ClassificationModelMetadata(TypedDict):
+    id: str
+    org_id: str
+    name: str
+    description: str | None
+    notes: str | None
+    version: int
+    memoryset_id: str
+    memory_lookup_count: int
+    storage_path: str
+    memoryset_collection_name: str
+    created_at: str
+    updated_at: str
+    locked: bool
+    num_classes: int
+    head_type: RACHeadType
+    weigh_memories: bool | None
+    min_memory_weight: float | None
+
+
+class ClassificationPredictionRequest(TypedDict):
+    input_values: list[str] | list[bytes] | list[str | bytes]
+    expected_labels: NotRequired[list[int] | None]
+    filters: NotRequired[list[FilterItem]]
+    tags: NotRequired[list[str]]
+    memoryset_override_name_or_id: NotRequired[str | None]
+    save_telemetry: NotRequired[bool]
+    save_telemetry_synchronously: NotRequired[bool]
+    prompt: NotRequired[str | None]
+    use_lookup_cache: NotRequired[bool]
+    consistency_level: NotRequired[Literal["Bounded", "Session", "Strong", "Eventual"] | None]
+    ignore_unlabeled: NotRequired[bool]
+
+
+class CloneMemorysetRequest(TypedDict):
+    name: str
+    description: NotRequired[str | None]
+    notes: NotRequired[str | None]
+    pretrained_embedding_model_name: NotRequired[PretrainedEmbeddingModelName | None]
+    finetuned_embedding_model_name_or_id: NotRequired[str | None]
+    max_seq_length_override: NotRequired[int | None]
+    prompt: NotRequired[str]
+
+
+class ColumnInfo(TypedDict):
+    name: str
+    type: ColumnType
+    enum_options: NotRequired[list[str] | None]
+    string_values: NotRequired[list[str] | None]
+    int_values: NotRequired[list[int] | None]
+    contains_nones: NotRequired[bool]
+
+
+class ConceptMetrics(TypedDict):
+    id: int
+    name: str
+    description: str | None
+    primary_label: int | None
+    memory_count: int
+    subconcepts: list[SubConceptMetrics]
+    label_percentages: list[LabelPercentage]
+
+
+class CreateClassificationModelRequest(TypedDict):
+    name: str
+    description: NotRequired[str | None]
+    notes: NotRequired[str | None]
+    memoryset_name: NotRequired[str | None]
+    memoryset_name_or_id: str
+    memory_lookup_count: NotRequired[int | None]
+    head_type: NotRequired[RACHeadType]
+    weigh_memories: NotRequired[bool | None]
+    min_memory_weight: NotRequired[float | None]
+    num_classes: NotRequired[int | None]
+
+
+class CreateMemorysetRequest(TypedDict):
+    name: str
+    description: NotRequired[str | None]
+    notes: NotRequired[str | None]
+    datasource_name_or_id: str
+    datasource_label_column: NotRequired[str | None]
+    datasource_score_column: NotRequired[str | None]
+    datasource_value_column: str
+    datasource_source_id_column: NotRequired[str | None]
+    remove_duplicates: NotRequired[bool]
+    pretrained_embedding_model_name: NotRequired[PretrainedEmbeddingModelName | None]
+    finetuned_embedding_model_name_or_id: NotRequired[str | None]
+    max_seq_length_override: NotRequired[int | None]
+    label_names: NotRequired[list[str] | None]
+    index_type: NotRequired[Literal["FLAT", "IVF_FLAT", "IVF_SQ8", "IVF_PQ", "HNSW", "DISKANN"]]
+    index_params: NotRequired[dict[str, int | float | str]]
+    prompt: NotRequired[str]
+    hidden: NotRequired[bool]
+    batch_size: NotRequired[int]
+    subsample: NotRequired[int | float | None]
+    memory_type: NotRequired[MemoryType]
+
+
+class CreateRegressionModelRequest(TypedDict):
+    name: str
+    description: NotRequired[str | None]
+    notes: NotRequired[str | None]
+    memoryset_name_or_id: str
+    memory_lookup_count: NotRequired[int | None]
+    head_type: NotRequired[RARHeadType]
+
+
+class DatasourceMetadata(TypedDict):
+    id: str
+    org_id: str
+    name: str
+    description: str | None
+    storage_path: str
+    length: int
+    columns: list[ColumnInfo]
+    created_at: str
+    updated_at: str
+
+
+class EmbeddingEvaluationResponse(TypedDict):
+    job_id: str
+    org_id: str
+    finetuned_embedding_model_id: str | None
+    pretrained_embedding_model_name: PretrainedEmbeddingModelName | None
+    datasource_id: str
+    subsample: int | float | None
+    datasource_value_column: str
+    datasource_label_column: NotRequired[str | None]
+    datasource_score_column: NotRequired[str | None]
+    neighbor_count: int
+    weigh_memories: bool
+    status: JobStatus
+    result: ClassificationMetrics | RegressionMetrics | None
+    created_at: str
+    updated_at: str
+    task_id: str
+
+
+class EvaluationResponse(TypedDict):
+    job_id: str
+    org_id: str
+    status: JobStatus
+    result: ClassificationMetrics | RegressionMetrics | None
+    created_at: str
+    updated_at: str
+    task_id: str
+
+
+class EvaluationResponseClassificationMetrics(TypedDict):
+    job_id: str
+    org_id: str
+    status: JobStatus
+    result: ClassificationMetrics | None
+    created_at: str
+    updated_at: str
+    task_id: str
+
+
+class EvaluationResponseRegressionMetrics(TypedDict):
+    job_id: str
+    org_id: str
+    status: JobStatus
+    result: RegressionMetrics | None
+    created_at: str
+    updated_at: str
+    task_id: str
+
+
+class FinetuneEmbeddingModelRequest(TypedDict):
+    name: str
+    base_model: PretrainedEmbeddingModelName
+    train_memoryset_name_or_id: NotRequired[str | None]
+    train_datasource_name_or_id: NotRequired[str | None]
+    eval_datasource_name_or_id: NotRequired[str | None]
+    label_column: NotRequired[str | None]
+    score_column: NotRequired[str | None]
+    value_column: NotRequired[str]
+    training_method: NotRequired[EmbeddingFinetuningMethod]
+    training_args: NotRequired[dict[str, str | int | float | bool]]
+
+
+class FinetunedEmbeddingModelMetadata(TypedDict):
+    embedding_dim: int
+    max_seq_length: int
+    uses_context: bool
+    id: str
+    org_id: str
+    name: str
+    storage_path: str
+    created_at: str
+    updated_at: str
+    base_model: PretrainedEmbeddingModelName
+    finetuning_job_id: str
+    finetuning_status: JobStatus
+    finetuning_task_id: str
+
+
+class HTTPValidationError(TypedDict):
+    detail: NotRequired[list[ValidationError]]
+
+
+class InvalidInputErrorResponse(TypedDict):
+    status_code: NotRequired[int]
+    validation_issues: list[FieldValidationError]
+
+
+class Job(TypedDict):
+    status: JobStatus
+    steps_total: int | None
+    steps_completed: int | None
+    exception: str | None
+    updated_at: str
+    created_at: str
+    id: str
+    org_id: str
+    worker_id: str | None
+    type: str
+    payload: BaseModel
+    result: BaseModel | None
+    depends_on: NotRequired[list[str]]
+    lease_token: str | None
+
+
+class LabelPredictionMemoryLookup(TypedDict):
+    value: str | bytes
+    embedding: list[float]
+    source_id: str | None
+    partition_id: str | None
+    metadata: dict[str, str | int | float | bool | None]
+    memory_id: str
+    memory_version: int
+    created_at: str
+    updated_at: str
+    edited_at: str
+    metrics: MemoryMetrics
+    label: int | None
+    label_name: str | None
+    lookup_score: float
+    prediction_id: str
+    attention_weight: float
+
+
+class LabelPredictionWithMemoriesAndFeedback(TypedDict):
+    prediction_id: str
+    confidence: float
+    anomaly_score: float | None
+    label: int | None
+    label_name: str | None
+    logits: list[float]
+    timestamp: str
+    input_value: str | bytes
+    input_embedding: list[float]
+    expected_label: int | None
+    expected_label_name: str | None
+    memories: list[LabelPredictionMemoryLookup]
+    org_id: str
+    memoryset_id: str
+    model_id: str
+    updated_at: str
+    tags: list[str]
+    explanation: str | None
+    memory_id: str | None
+    feedbacks: list[PredictionFeedback]
+
+
+class LabeledMemory(TypedDict):
+    value: str | bytes
+    embedding: list[float]
+    source_id: str | None
+    partition_id: str | None
+    metadata: dict[str, str | int | float | bool | None]
+    memory_id: str
+    memory_version: int
+    created_at: str
+    updated_at: str
+    edited_at: str
+    metrics: MemoryMetrics
+    label: int | None
+    label_name: str | None
+
+
+class LabeledMemoryLookup(TypedDict):
+    value: str | bytes
+    embedding: list[float]
+    source_id: str | None
+    partition_id: str | None
+    metadata: dict[str, str | int | float | bool | None]
+    memory_id: str
+    memory_version: int
+    created_at: str
+    updated_at: str
+    edited_at: str
+    metrics: MemoryMetrics
+    label: int | None
+    label_name: str | None
+    lookup_score: float
+
+
+class LabeledMemoryUpdate(TypedDict):
+    memory_id: str
+    value: NotRequired[str | bytes]
+    metadata: NotRequired[dict[str, str | int | float | bool | None] | None]
+    source_id: NotRequired[str | None]
+    partition_id: NotRequired[str | None]
+    metrics: NotRequired[MemoryMetrics | None]
+    label: NotRequired[int | None]
+
+
+class LabeledMemoryWithFeedbackMetrics(TypedDict):
+    value: str | bytes
+    embedding: list[float]
+    source_id: str | None
+    partition_id: str | None
+    metadata: dict[str, str | int | float | bool | None]
+    memory_id: str
+    memory_version: int
+    created_at: str
+    updated_at: str
+    edited_at: str
+    metrics: MemoryMetrics
+    label: int | None
+    label_name: str | None
+    feedback_metrics: dict[str, FeedbackMetrics]
+    lookup_count: int
+
+
+class ListPredictionsRequest(TypedDict):
+    model_id: NotRequired[str | None]
+    tag: NotRequired[str | None]
+    prediction_ids: NotRequired[list[str] | None]
+    start_timestamp: NotRequired[str | None]
+    end_timestamp: NotRequired[str | None]
+    memory_id: NotRequired[str | None]
+    limit: NotRequired[int]
+    offset: NotRequired[int | None]
+    sort: NotRequired[PredictionSort]
+    expected_label_match: NotRequired[bool | None]
+
+
+class MemorysetAnalysisConfigs(TypedDict):
+    distribution: NotRequired[MemorysetDistributionAnalysisConfig | None]
+    label: NotRequired[MemorysetLabelAnalysisConfig | None]
+    duplicate: NotRequired[MemorysetDuplicateAnalysisConfig | None]
+    projection: NotRequired[MemorysetProjectionAnalysisConfig | None]
+    cluster: NotRequired[MemorysetClusterAnalysisConfig | None]
+    class_patterns: NotRequired[MemorysetClassPatternsAnalysisConfig | None]
+    concepts: NotRequired[MemorysetConceptAnalysisConfig | None]
+
+
+class MemorysetAnalysisRequest(TypedDict):
+    lookup_count: NotRequired[int]
+    batch_size: NotRequired[int]
+    clear_metrics: NotRequired[bool]
+    configs: MemorysetAnalysisConfigs
+
+
+class MemorysetConceptMetrics(TypedDict):
+    concepts: list[ConceptMetrics]
+    num_outliers: int
+    updated_at: str
+
+
+class MemorysetMetrics(TypedDict):
+    distribution: NotRequired[MemorysetDistributionMetrics | None]
+    label: NotRequired[MemorysetLabelMetrics | None]
+    duplicate: NotRequired[MemorysetDuplicateMetrics | None]
+    projection: NotRequired[MemorysetProjectionMetrics | None]
+    cluster: NotRequired[MemorysetClusterMetrics | None]
+    class_patterns: NotRequired[MemorysetClassPatternsMetrics | None]
+    concepts: NotRequired[MemorysetConceptMetrics | None]
+
+
+class PaginatedJob(TypedDict):
+    items: list[Job]
+    total: int
+    offset: int
+    limit: int
+
+
+class PaginatedUnionLabeledMemoryWithFeedbackMetricsScoredMemoryWithFeedbackMetrics(TypedDict):
+    items: list[LabeledMemoryWithFeedbackMetrics | ScoredMemoryWithFeedbackMetrics]
+    total: int
+    offset: int
+    limit: int
+
+
+class PretrainedEmbeddingModelMetadata(TypedDict):
+    embedding_dim: int
+    max_seq_length: int
+    uses_context: bool
+    name: PretrainedEmbeddingModelName
+    experimental: NotRequired[bool]
+    supports_instructions: bool
+    num_params: int
+
+
+class TelemetryMemoriesRequest(TypedDict):
+    memoryset_id: str
+    offset: NotRequired[int]
+    limit: NotRequired[int]
+    filters: NotRequired[list[FilterItem | TelemetryFilterItem]]
+    sort: NotRequired[list[TelemetrySortOptions] | None]
+
+
+class WorkerInfo(TypedDict):
+    id: str
+    status: WorkerStatus
+    started_at: str
+    updated_at: str
+    version: str | None
+    git_sha: str
+    config: dict[str, str | float | int | bool | dict[str, str] | None]
+
+
+PatchGpuMemorysetByNameOrIdMemoryRequest = LabeledMemoryUpdate | ScoredMemoryUpdate
+
+
+PatchGpuMemorysetByNameOrIdMemoriesRequest = list[LabeledMemoryUpdate] | list[ScoredMemoryUpdate]
+
+
+class CascadingEditSuggestion(TypedDict):
+    neighbor: LabeledMemoryLookup
+    suggested_label: int
+    lookup_score: float
+
+
+class MemorysetAnalysisResponse(TypedDict):
+    job_id: str
+    org_id: str
+    memoryset_id: str
+    status: JobStatus
+    lookup_count: int
+    batch_size: int
+    clear_metrics: bool
+    configs: MemorysetAnalysisConfigs
+    results: MemorysetMetrics | None
+    created_at: str
+    updated_at: str
+    task_id: str
+
+
+class MemorysetMetadata(TypedDict):
+    id: str
+    org_id: str
+    collection_name: str
+    name: str
+    description: str | None
+    notes: str | None
+    length: int
+    pretrained_embedding_model_name: PretrainedEmbeddingModelName | None
+    finetuned_embedding_model_id: str | None
+    created_at: str
+    updated_at: str
+    memories_updated_at: str
+    insertion_job_id: str
+    insertion_status: JobStatus
+    metrics: MemorysetMetrics
+    memory_type: MemoryType
+    label_names: list[str] | None
+    index_type: Literal["FLAT", "IVF_FLAT", "IVF_SQ8", "IVF_PQ", "HNSW", "DISKANN"]
+    index_params: dict[str, Any]
+    database_uri: str | None
+    document_prompt_override: str | None
+    query_prompt_override: str | None
+    hidden: bool
+    insertion_task_id: str
+
+
+class PaginatedWorkerInfo(TypedDict):
+    items: list[WorkerInfo]
+    total: int
+    offset: int
+    limit: int
+
+
+class BootstrapClassificationModelMeta(TypedDict):
+    datasource_meta: DatasourceMetadata
+    memoryset_meta: MemorysetMetadata
+    model_meta: ClassificationModelMetadata
+    agent_output: BootstrapClassificationModelResult
+
+
+class BootstrapClassificationModelResponse(TypedDict):
+    job_id: str
+    org_id: str
+    status: JobStatus
+    result: BootstrapClassificationModelMeta | None
+    input: BootstrapClassificationModelRequest | None
+    task_id: str
+
+
+class OrcaAsyncClient(AsyncClient):
+    @staticmethod
+    def _parse_params(
+        params: Mapping[str, Any],
+        path: str,
+    ) -> tuple[dict[str, Any], dict[str, Any]]:
+        placeholders = {name for _, name, _, _ in Formatter().parse(path) if name}
+        path_params = {k: v for k, v in params.items() if k in placeholders}
+        query_params = {k: v for k, v in params.items() if k not in placeholders and v is not None}
+        if placeholders - path_params.keys():
+            raise ValueError(f"Missing path params: {', '.join(placeholders - path_params.keys())}")
+        return path_params, query_params
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/test_error/{status_code}"],
+        *,
+        params: GetTestErrorByStatusCodeParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> Any:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/check/alive"],
+        *,
+        params: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> AliveResponse:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/check/ready"],
+        *,
+        params: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> ReadyResponse:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/gpu/check/healthy"],
+        *,
+        params: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> HealthyResponse:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/check/healthy"],
+        *,
+        params: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> HealthyResponse:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/gpu/config"],
+        *,
+        params: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> dict[str, str | float | int | bool | None]:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/config"],
+        *,
+        params: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> dict[str, str | float | int | bool | None]:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/auth/root"],
+        *,
+        params: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> bool:
+        """Return true only when called with a valid root API key; otherwise 401 Unauthenticated."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/auth/api_key"],
+        *,
+        params: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[ApiKeyMetadata]:
+        """List all API keys for the organization."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/auth"],
+        *,
+        params: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> bool:
+        """Returns true if the api key header is valid for the org (will be false for admin api key)"""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/auth/org/plan"],
+        *,
+        params: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> OrgPlan:
+        """Get the organization plan."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/memoryset"],
+        *,
+        params: GetMemorysetParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[MemorysetMetadata]:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/memoryset/{name_or_id}"],
+        *,
+        params: GetMemorysetByNameOrIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> MemorysetMetadata:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/memoryset/{name_or_id}/memory/{memory_id}"],
+        *,
+        params: GetMemorysetByNameOrIdMemoryByMemoryIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> LabeledMemory | ScoredMemory:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/memoryset/{name_or_id}/potential_duplicate_groups"],
+        *,
+        params: GetMemorysetByNameOrIdPotentialDuplicateGroupsParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[list[LabeledMemory]] | list[list[ScoredMemory]]:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/memoryset/{name_or_id}/analysis"],
+        *,
+        params: GetMemorysetByNameOrIdAnalysisParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[MemorysetAnalysisResponse]:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/memoryset/{name_or_id}/analysis/{analysis_job_id}"],
+        *,
+        params: GetMemorysetByNameOrIdAnalysisByAnalysisJobIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> MemorysetAnalysisResponse:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/finetuned_embedding_model"],
+        *,
+        params: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[FinetunedEmbeddingModelMetadata]:
+        """List all finetuned embedding models for the organization."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/finetuned_embedding_model/{name_or_id}"],
+        *,
+        params: GetFinetunedEmbeddingModelByNameOrIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> FinetunedEmbeddingModelMetadata:
+        """Get a finetuned embedding model by name or ID."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/pretrained_embedding_model"],
+        *,
+        params: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[PretrainedEmbeddingModelMetadata]:
+        """List all available pretrained embedding models."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/pretrained_embedding_model/{model_name}"],
+        *,
+        params: GetPretrainedEmbeddingModelByModelNameParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> PretrainedEmbeddingModelMetadata:
+        """Get metadata for a specific pretrained embedding model."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/finetuned_embedding_model/{name_or_id}/evaluation/{job_id}"],
+        *,
+        params: GetFinetunedEmbeddingModelByNameOrIdEvaluationByJobIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> EmbeddingEvaluationResponse:
+        """Get evaluation results for a finetuned embedding model by job ID."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/pretrained_embedding_model/{model_name}/evaluation/{job_id}"],
+        *,
+        params: GetPretrainedEmbeddingModelByModelNameEvaluationByJobIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> EmbeddingEvaluationResponse:
+        """Get evaluation results for a pretrained embedding model by job ID."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/finetuned_embedding_model/{name_or_id}/evaluations"],
+        *,
+        params: GetFinetunedEmbeddingModelByNameOrIdEvaluationsParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[EmbeddingEvaluationResponse]:
+        """List all evaluation results for a finetuned embedding model."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/pretrained_embedding_model/{model_name}/evaluations"],
+        *,
+        params: GetPretrainedEmbeddingModelByModelNameEvaluationsParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[EmbeddingEvaluationResponse]:
+        """List all evaluation results for a pretrained embedding model."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/datasource"],
+        *,
+        params: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[DatasourceMetadata]:
+        """List all datasources for the organization."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/datasource/{name_or_id}"],
+        *,
+        params: GetDatasourceByNameOrIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> DatasourceMetadata:
+        """Get a datasource by name or ID."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/datasource/{name_or_id}/embedding_model_evaluations"],
+        *,
+        params: GetDatasourceByNameOrIdEmbeddingModelEvaluationsParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[EmbeddingEvaluationResponse]:
+        """List all evaluation results for a datasource."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/datasource/{name_or_id}/download"],
+        *,
+        params: GetDatasourceByNameOrIdDownloadParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[dict[str, Any]]:
+        """Download datasource in the specified format."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/datasource/{name_or_id}/download"],
+        *,
+        params: GetDatasourceByNameOrIdDownloadParams,
+        parse_as: Literal["text"],
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> str:
+        """Download datasource in the specified format."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/datasource/{name_or_id}/download"],
+        *,
+        params: GetDatasourceByNameOrIdDownloadParams,
+        parse_as: None,
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> bytes:
+        """Download datasource in the specified format."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/classification_model"],
+        *,
+        params: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[ClassificationModelMetadata]:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/regression_model"],
+        *,
+        params: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[RegressionModelMetadata]:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/classification_model/{name_or_id}"],
+        *,
+        params: GetClassificationModelByNameOrIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> ClassificationModelMetadata:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/regression_model/{name_or_id}"],
+        *,
+        params: GetRegressionModelByNameOrIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> RegressionModelMetadata:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/predictive_model"],
+        *,
+        params: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[ClassificationModelMetadata | RegressionModelMetadata]:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/classification_model/{model_name_or_id}/evaluation"],
+        *,
+        params: GetClassificationModelByModelNameOrIdEvaluationParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[EvaluationResponseClassificationMetrics]:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/regression_model/{model_name_or_id}/evaluation"],
+        *,
+        params: GetRegressionModelByModelNameOrIdEvaluationParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[EvaluationResponseRegressionMetrics]:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/classification_model/{model_name_or_id}/evaluation/{job_id}"],
+        *,
+        params: GetClassificationModelByModelNameOrIdEvaluationByJobIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> EvaluationResponseClassificationMetrics:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/regression_model/{model_name_or_id}/evaluation/{job_id}"],
+        *,
+        params: GetRegressionModelByModelNameOrIdEvaluationByJobIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> EvaluationResponseRegressionMetrics:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/job/{job_id}"],
+        *,
+        params: GetJobByJobIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> Job:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/job/{job_id}/status"],
+        *,
+        params: GetJobByJobIdStatusParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> JobStatusInfo:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/job"],
+        *,
+        params: GetJobParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> PaginatedJob:
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/worker"],
+        *,
+        params: GetWorkerParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> PaginatedWorkerInfo:
+        """
+        List all workers in the system. Requires root access.
+
+        This endpoint automatically cleans up orphaned workers before returning results.
+        """
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/worker/{worker_id}"],
+        *,
+        params: GetWorkerByWorkerIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> WorkerInfo:
+        """Get information about a specific worker. Requires root access."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/telemetry/prediction/{prediction_id}"],
+        *,
+        params: GetTelemetryPredictionByPredictionIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> LabelPredictionWithMemoriesAndFeedback | ScorePredictionWithMemoriesAndFeedback:
+        """Get a specific prediction by ID."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/telemetry/prediction/{prediction_id}/explanation"],
+        *,
+        params: GetTelemetryPredictionByPredictionIdExplanationParams,
+        parse_as: Literal["text"],
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> str:
+        """Get explanation for a prediction, optionally streaming the response."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/telemetry/prediction/{prediction_id}/action"],
+        *,
+        params: GetTelemetryPredictionByPredictionIdActionParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> ActionRecommendation:
+        """Get action recommendation for improving a specific prediction."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/telemetry/prediction/{prediction_id}/memory_suggestions"],
+        *,
+        params: GetTelemetryPredictionByPredictionIdMemorySuggestionsParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> AddMemoryRecommendations:
+        """
+        Generate synthetic memory suggestions to improve a specific prediction.
+
+        The returned suggestions have labels as string representations of integer indices
+        corresponding to the memoryset's label_names.
+        """
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/telemetry/feedback_category"],
+        *,
+        params: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[PredictionFeedbackCategory]:
+        """List all feedback categories for the organization."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/telemetry/feedback_category/{name_or_id}"],
+        *,
+        params: GetTelemetryFeedbackCategoryByNameOrIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> PredictionFeedbackCategory:
+        """Get a feedback category by name or ID."""
+        pass
+
+    @overload
+    async def GET(
+        self,
+        path: Literal["/agents/bootstrap_classification_model/{job_id}"],
+        *,
+        params: GetAgentsBootstrapClassificationModelByJobIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> BootstrapClassificationModelResponse:
+        """Get the status of a bootstrap classification model job"""
+        pass
+
+    async def GET(
+        self,
+        path: str,
+        *,
+        params: Mapping[str, Any] | None = None,
+        parse_as: Literal["json", "text"] | None = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> Any:
+        path_params, query_params = self._parse_params(params or {}, path)
+        res = await self.get(
+            path.format(**path_params),
+            params=query_params,
+            headers=headers,
+            cookies=cookies,
+            auth=auth,
+            follow_redirects=follow_redirects,
+            timeout=timeout,
+            extensions=extensions,
+        )
+        res.raise_for_status()
+        return (
+            None
+            if res.status_code == 204
+            else res.json() if parse_as == "json" else res.text if parse_as == "text" else res.content
+        )
+
+    @overload
+    async def DELETE(
+        self,
+        path: Literal["/cleanup"],
+        *,
+        params: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> CleanupResponse:
+        """Cleanup orphaned milvus collections and blobs"""
+        pass
+
+    @overload
+    async def DELETE(
+        self,
+        path: Literal["/auth/api_key/{name_or_id}"],
+        *,
+        params: DeleteAuthApiKeyByNameOrIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> None:
+        """Delete an API key by name or ID."""
+        pass
+
+    @overload
+    async def DELETE(
+        self,
+        path: Literal["/auth/org"],
+        *,
+        params: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> None:
+        """Deletes the org and all associated resources"""
+        pass
+
+    @overload
+    async def DELETE(
+        self,
+        path: Literal["/memoryset/{name_or_id}"],
+        *,
+        params: DeleteMemorysetByNameOrIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> None:
+        pass
+
+    @overload
+    async def DELETE(
+        self,
+        path: Literal["/memoryset/{name_or_id}/memory/{memory_id}"],
+        *,
+        params: DeleteMemorysetByNameOrIdMemoryByMemoryIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> None:
+        pass
+
+    @overload
+    async def DELETE(
+        self,
+        path: Literal["/finetuned_embedding_model/{name_or_id}"],
+        *,
+        params: DeleteFinetunedEmbeddingModelByNameOrIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> None:
+        """Delete a finetuned embedding model by name or ID."""
+        pass
+
+    @overload
+    async def DELETE(
+        self,
+        path: Literal["/datasource/{name_or_id}"],
+        *,
+        params: DeleteDatasourceByNameOrIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> None:
+        """Delete a datasource by name or ID."""
+        pass
+
+    @overload
+    async def DELETE(
+        self,
+        path: Literal["/classification_model/{name_or_id}"],
+        *,
+        params: DeleteClassificationModelByNameOrIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> None:
+        pass
+
+    @overload
+    async def DELETE(
+        self,
+        path: Literal["/regression_model/{name_or_id}"],
+        *,
+        params: DeleteRegressionModelByNameOrIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> None:
+        pass
+
+    @overload
+    async def DELETE(
+        self,
+        path: Literal["/classification_model/{model_name_or_id}/evaluation/{job_id}"],
+        *,
+        params: DeleteClassificationModelByModelNameOrIdEvaluationByJobIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> None:
+        pass
+
+    @overload
+    async def DELETE(
+        self,
+        path: Literal["/regression_model/{model_name_or_id}/evaluation/{job_id}"],
+        *,
+        params: DeleteRegressionModelByModelNameOrIdEvaluationByJobIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> None:
+        pass
+
+    @overload
+    async def DELETE(
+        self,
+        path: Literal["/job/{job_id}/abort"],
+        *,
+        params: DeleteJobByJobIdAbortParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> None:
+        pass
+
+    @overload
+    async def DELETE(
+        self,
+        path: Literal["/telemetry/feedback_category/{name_or_id}"],
+        *,
+        params: DeleteTelemetryFeedbackCategoryByNameOrIdParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> None:
+        """Delete a feedback category and all associated feedback records."""
+        pass
+
+    async def DELETE(
+        self,
+        path: str,
+        *,
+        params: Mapping[str, Any] | None = None,
+        parse_as: Literal["json", "text"] | None = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> Any:
+        path_params, query_params = self._parse_params(params or {}, path)
+        res = await self.delete(
+            path.format(**path_params),
+            params=query_params,
+            headers=headers,
+            cookies=cookies,
+            auth=auth,
+            follow_redirects=follow_redirects,
+            timeout=timeout,
+            extensions=extensions,
+        )
+        res.raise_for_status()
+        return (
+            None
+            if res.status_code == 204
+            else res.json() if parse_as == "json" else res.text if parse_as == "text" else res.content
+        )
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/auth/api_key"],
+        *,
+        params: None = None,
+        json: CreateApiKeyRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> CreateApiKeyResponse:
+        """Create a new API key for the organization."""
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/auth/org/plan"],
+        *,
+        params: None = None,
+        json: CreateOrgPlanRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> OrgPlan:
+        """Create an organization plan."""
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/memoryset"],
+        *,
+        params: None = None,
+        json: CreateMemorysetRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> MemorysetMetadata:
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/memoryset/{name_or_id}/clone"],
+        *,
+        params: PostMemorysetByNameOrIdCloneParams,
+        json: CloneMemorysetRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> MemorysetMetadata:
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/batch_delete_memoryset"],
+        *,
+        params: None = None,
+        json: DeleteMemorysetsRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> None:
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/gpu/memoryset/{name_or_id}/lookup"],
+        *,
+        params: PostGpuMemorysetByNameOrIdLookupParams,
+        json: LookupRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[list[LabeledMemoryLookup | ScoredMemoryLookup]]:
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/memoryset/{name_or_id}/memories/get"],
+        *,
+        params: PostMemorysetByNameOrIdMemoriesGetParams,
+        json: GetMemoriesRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[LabeledMemory] | list[ScoredMemory]:
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/memoryset/{name_or_id}/memories"],
+        *,
+        params: PostMemorysetByNameOrIdMemoriesParams,
+        json: ListMemoriesRequest | None = None,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[LabeledMemory] | list[ScoredMemory]:
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/memoryset/{name_or_id}/memories/delete"],
+        *,
+        params: PostMemorysetByNameOrIdMemoriesDeleteParams,
+        json: DeleteMemoriesRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> None:
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/gpu/memoryset/{name_or_id}/memory"],
+        *,
+        params: PostGpuMemorysetByNameOrIdMemoryParams,
+        json: PostGpuMemorysetByNameOrIdMemoryRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[str]:
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/memoryset/{name_or_id}/analysis"],
+        *,
+        params: PostMemorysetByNameOrIdAnalysisParams,
+        json: MemorysetAnalysisRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> MemorysetAnalysisResponse:
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/memoryset/{name_or_id}/memory/{memory_id}/cascading_edits"],
+        *,
+        params: PostMemorysetByNameOrIdMemoryByMemoryIdCascadingEditsParams,
+        json: CascadeEditSuggestionsRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[CascadingEditSuggestion]:
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/finetuned_embedding_model"],
+        *,
+        params: None = None,
+        json: FinetuneEmbeddingModelRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> FinetunedEmbeddingModelMetadata:
+        """Create a finetuned embedding model."""
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/gpu/finetuned_embedding_model/{name_or_id}/embedding"],
+        *,
+        params: PostGpuFinetunedEmbeddingModelByNameOrIdEmbeddingParams,
+        json: EmbedRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[list[float]]:
+        """Embed values using a finetuned embedding model."""
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/gpu/pretrained_embedding_model/{model_name}/embedding"],
+        *,
+        params: PostGpuPretrainedEmbeddingModelByModelNameEmbeddingParams,
+        json: EmbedRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[list[float]]:
+        """Embed values using a pretrained embedding model."""
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/finetuned_embedding_model/{name_or_id}/evaluation"],
+        *,
+        params: PostFinetunedEmbeddingModelByNameOrIdEvaluationParams,
+        json: EmbeddingEvaluationRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> EmbeddingEvaluationResponse:
+        """Evaluate a finetuned embedding model as a KNN classifier or regressor."""
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/pretrained_embedding_model/{model_name}/evaluation"],
+        *,
+        params: PostPretrainedEmbeddingModelByModelNameEvaluationParams,
+        json: EmbeddingEvaluationRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> EmbeddingEvaluationResponse:
+        """Evaluate a pretrained embedding model as a KNN classifier or regressor."""
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/datasource/upload"],
+        *,
+        params: None = None,
+        json: None = None,
+        data: PostDatasourceUploadRequest,
+        files: dict[Literal["files"], FileTypes] | list[tuple[Literal["files"], FileTypes]],
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> DatasourceMetadata:
+        """
+        Create a datasource by uploading files.
+
+        Supports multiple file upload scenarios:
+        - Multiple files: HuggingFace Dataset format (dataset_info.json, state.json, .arrow files, etc.)
+        - Single file: CSV, JSON, JSONL, Parquet, or Pickle files
+        """
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/datasource"],
+        *,
+        params: None = None,
+        json: CreateDatasourceFromContentRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> DatasourceMetadata:
+        """
+        Create a datasource from JSON content.
+
+        Automatically detects and supports multiple JSON formats:
+        - List of records: [{"col1": val1, "col2": val2}, {"col1": val3, "col2": val4}, ...]
+        - Dictionary of columns: {"col1": [val1, val3, ...], "col2": [val2, val4, ...]}
+        """
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/datasource/{name_or_id}/rows"],
+        *,
+        params: PostDatasourceByNameOrIdRowsParams,
+        json: GetDatasourceRowsRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[dict[str, Any]]:
+        """Get rows from a specific datasource with optional filtering."""
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/datasource/{name_or_id}/rows/count"],
+        *,
+        params: PostDatasourceByNameOrIdRowsCountParams,
+        json: GetDatasourceRowCountRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> int:
+        """Get row count from a specific datasource with optional filtering."""
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/classification_model"],
+        *,
+        params: None = None,
+        json: CreateClassificationModelRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> ClassificationModelMetadata:
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/regression_model"],
+        *,
+        params: None = None,
+        json: CreateRegressionModelRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> RegressionModelMetadata:
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/gpu/classification_model/{name_or_id}/prediction"],
+        *,
+        params: PostGpuClassificationModelByNameOrIdPredictionParams,
+        json: ClassificationPredictionRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[BaseLabelPredictionResult]:
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/classification_model/{name_or_id}/prediction"],
+        *,
+        params: PostClassificationModelByNameOrIdPredictionParams,
+        json: ClassificationPredictionRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[BaseLabelPredictionResult]:
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/gpu/regression_model/{name_or_id}/prediction"],
+        *,
+        params: PostGpuRegressionModelByNameOrIdPredictionParams,
+        json: RegressionPredictionRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[BaseScorePredictionResult]:
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/regression_model/{name_or_id}/prediction"],
+        *,
+        params: PostRegressionModelByNameOrIdPredictionParams,
+        json: RegressionPredictionRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[BaseScorePredictionResult]:
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/classification_model/{model_name_or_id}/evaluation"],
+        *,
+        params: PostClassificationModelByModelNameOrIdEvaluationParams,
+        json: ClassificationEvaluationRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> EvaluationResponse:
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/regression_model/{model_name_or_id}/evaluation"],
+        *,
+        params: PostRegressionModelByModelNameOrIdEvaluationParams,
+        json: RegressionEvaluationRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> EvaluationResponse:
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/telemetry/prediction"],
+        *,
+        params: None = None,
+        json: ListPredictionsRequest | None = None,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[LabelPredictionWithMemoriesAndFeedback | ScorePredictionWithMemoriesAndFeedback]:
+        """List predictions with optional filtering and sorting."""
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/telemetry/prediction/count"],
+        *,
+        params: None = None,
+        json: CountPredictionsRequest | None = None,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> int:
+        """Count predictions with optional filtering."""
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/telemetry/memories"],
+        *,
+        params: None = None,
+        json: TelemetryMemoriesRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> PaginatedUnionLabeledMemoryWithFeedbackMetricsScoredMemoryWithFeedbackMetrics:
+        """
+        List memories with feedback metrics.
+        **Note**: This endpoint will ONLY return memories that have been used in a prediction.
+        If you want to query ALL memories WITHOUT feedback metrics, use the query_memoryset endpoint.
+        """
+        pass
+
+    @overload
+    async def POST(
+        self,
+        path: Literal["/agents/bootstrap_classification_model"],
+        *,
+        params: None = None,
+        json: BootstrapClassificationModelRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> BootstrapClassificationModelResponse:
+        """
+        Bootstrap a classification model by creating a memoryset with generated memories and a classification model.
+
+        This endpoint uses the bootstrap_classification_model agent to generate:
+        1. Memoryset configuration with appropriate settings
+        2. Model configuration with optimal parameters
+        3. High-quality training memories for each label
+
+        The process involves:
+        1. Calling the agent to generate configurations and memories
+        2. Creating a datasource from the generated memories
+        3. Creating a memoryset from the datasource
+        4. Creating a classification model from the memoryset
+        """
+        pass
+
+    async def POST(
+        self,
+        path: str,
+        *,
+        params: Mapping[str, Any] | None = None,
+        json: Any | None = None,
+        data: Mapping[str, Any] | None = None,
+        content: RequestContent | None = None,
+        files: dict[Any, FileTypes] | list[tuple[Any, FileTypes]] | None = None,
+        parse_as: Literal["json", "text"] | None = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> Any:
+        path_params, query_params = self._parse_params(params or {}, path)
+        res = await self.post(
+            path.format(**path_params),
+            params=query_params,
+            content=content,
+            data=data,
+            files=files,
+            json=json,
+            headers=headers,
+            cookies=cookies,
+            auth=auth,
+            follow_redirects=follow_redirects,
+            timeout=timeout,
+            extensions=extensions,
+        )
+        res.raise_for_status()
+        return (
+            None
+            if res.status_code == 204
+            else res.json() if parse_as == "json" else res.text if parse_as == "text" else res.content
+        )
+
+    @overload
+    async def PUT(
+        self,
+        path: Literal["/auth/org/plan"],
+        *,
+        params: None = None,
+        json: UpdateOrgPlanRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> OrgPlan:
+        """Update the organization plan."""
+        pass
+
+    @overload
+    async def PUT(
+        self,
+        path: Literal["/telemetry/prediction/feedback"],
+        *,
+        params: None = None,
+        json: PutTelemetryPredictionFeedbackRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> PredictionFeedbackResult:
+        """Record feedback for predictions, handling updates, deletions, and insertions."""
+        pass
+
+    async def PUT(
+        self,
+        path: str,
+        *,
+        params: Mapping[str, Any] | None = None,
+        json: Any | None = None,
+        data: Mapping[str, Any] | None = None,
+        content: RequestContent | None = None,
+        files: dict[Any, FileTypes] | list[tuple[Any, FileTypes]] | None = None,
+        parse_as: Literal["json", "text"] | None = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> Any:
+        path_params, query_params = self._parse_params(params or {}, path)
+        res = await self.put(
+            path.format(**path_params),
+            params=query_params,
+            content=content,
+            data=data,
+            files=files,
+            json=json,
+            headers=headers,
+            cookies=cookies,
+            auth=auth,
+            follow_redirects=follow_redirects,
+            timeout=timeout,
+            extensions=extensions,
+        )
+        res.raise_for_status()
+        return (
+            None
+            if res.status_code == 204
+            else res.json() if parse_as == "json" else res.text if parse_as == "text" else res.content
+        )
+
+    @overload
+    async def PATCH(
+        self,
+        path: Literal["/memoryset/{name_or_id}"],
+        *,
+        params: PatchMemorysetByNameOrIdParams,
+        json: MemorysetUpdate,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> MemorysetMetadata:
+        pass
+
+    @overload
+    async def PATCH(
+        self,
+        path: Literal["/gpu/memoryset/{name_or_id}/memory"],
+        *,
+        params: PatchGpuMemorysetByNameOrIdMemoryParams,
+        json: PatchGpuMemorysetByNameOrIdMemoryRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> LabeledMemory | ScoredMemory:
+        pass
+
+    @overload
+    async def PATCH(
+        self,
+        path: Literal["/gpu/memoryset/{name_or_id}/memories"],
+        *,
+        params: PatchGpuMemorysetByNameOrIdMemoriesParams,
+        json: PatchGpuMemorysetByNameOrIdMemoriesRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[LabeledMemory] | list[ScoredMemory]:
+        pass
+
+    @overload
+    async def PATCH(
+        self,
+        path: Literal["/classification_model/{name_or_id}"],
+        *,
+        params: PatchClassificationModelByNameOrIdParams,
+        json: PredictiveModelUpdate,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> ClassificationModelMetadata:
+        pass
+
+    @overload
+    async def PATCH(
+        self,
+        path: Literal["/regression_model/{name_or_id}"],
+        *,
+        params: PatchRegressionModelByNameOrIdParams,
+        json: PredictiveModelUpdate,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> RegressionModelMetadata:
+        pass
+
+    @overload
+    async def PATCH(
+        self,
+        path: Literal["/telemetry/prediction/{prediction_id}"],
+        *,
+        params: PatchTelemetryPredictionByPredictionIdParams,
+        json: UpdatePredictionRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> Any:
+        """Update a prediction with new expected values, tags, or memory ID."""
+        pass
+
+    async def PATCH(
+        self,
+        path: str,
+        *,
+        params: Mapping[str, Any] | None = None,
+        json: Any | None = None,
+        data: Mapping[str, Any] | None = None,
+        content: RequestContent | None = None,
+        files: dict[Any, FileTypes] | list[tuple[Any, FileTypes]] | None = None,
+        parse_as: Literal["json", "text"] | None = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> Any:
+        path_params, query_params = self._parse_params(params or {}, path)
+        res = await self.patch(
+            path.format(**path_params),
+            params=query_params,
+            content=content,
+            data=data,
+            files=files,
+            json=json,
+            headers=headers,
+            cookies=cookies,
+            auth=auth,
+            follow_redirects=follow_redirects,
+            timeout=timeout,
+            extensions=extensions,
+        )
+        res.raise_for_status()
+        return (
+            None
+            if res.status_code == 204
+            else res.json() if parse_as == "json" else res.text if parse_as == "text" else res.content
+        )
+
+    @staticmethod
+    async def _raise_error_for_response(response: Response) -> None:
+        if response.status_code == 401:
+            raise ValueError("Invalid API key")
+        # elif response.status_code == 402:
+        #     res = cast(QuotaExceededErrorResponse, json.loads((await response.aread()).decode(response.encoding or "utf-8")))
+
+        #     raise RuntimeError(
+        #         f"{res['quota_type'].replace('_', ' ').title()} limit reached ({res['current']}/{res['quota_limit']})"
+        #     )
+        elif response.status_code == 403:
+            raise PermissionError(json.loads((await response.aread()).decode(response.encoding or "utf-8"))["reason"])
+        elif response.status_code == 404:
+            res = cast(NotFoundErrorResponse, json.loads((await response.aread()).decode(response.encoding or "utf-8")))
+            if res["resource"] is not None:
+                raise LookupError(f"The {res['resource']} you are looking for does not exist")
+            else:
+                raise RuntimeError(f"Unknown API route: {response.url}")
+        elif response.status_code == 405:
+            raise RuntimeError(f"Unknown method {response.request.method} for API route: {response.url}")
+        elif response.status_code == 409:
+            res = cast(
+                ConstraintViolationErrorResponse,
+                json.loads((await response.aread()).decode(response.encoding or "utf-8")),
+            )
+            raise RuntimeError(res["constraint"])
+        elif response.status_code == 422:
+            res = cast(
+                InvalidInputErrorResponse, json.loads((await response.aread()).decode(response.encoding or "utf-8"))
+            )
+            issues = [f"{issue['loc'][-1]}: {issue['msg']}" for issue in res["validation_issues"]]
+            raise ValueError("Invalid input:\n\t" + "\n\t".join(issues))
+        elif response.status_code == 500:
+            res = cast(
+                InternalServerErrorResponse, json.loads((await response.aread()).decode(response.encoding or "utf-8"))
+            )
+            raise RuntimeError(f"Unexpected server error: {res['message']}")
+        elif response.status_code == 503:
+            raise RuntimeError("Orca API is currently unavailable, please try again later")
+        elif response.status_code >= 400:
+            raise RuntimeError(f"Unexpected status code: {response.status_code}")
+
+    @staticmethod
+    async def _instrument_request(request: Request) -> None:
+        request.headers["X-Request-ID"] = str(uuid.uuid4())
+
+    def __init__(
+        self,
+        *,
+        api_key: str | None = None,
+        base_url: URL | str = "",
+        headers: HeaderTypes | None = None,
+        transport: AsyncBaseTransport | None = None,
+        timeout: TimeoutTypes | None = None,
+        limits: Limits | None = None,
+        max_redirects: int = 20,
+        event_hooks: None | (Mapping[str, list[Callable[..., Any]]]) = None,
+        http1: bool = True,
+        http2: bool = False,
+        proxy: str | URL | Proxy | None = None,
+        log_level: int = logging.WARNING,
+    ) -> None:
+        """
+        Initialize an OrcaAPI async httpx client
+
+        Params:
+            api_key: API key to use for authentication, will default to ORCA_API_KEY if not set.
+            base_url: URL of the OrcaAPI, will default to ORCA_API_URL or the cloud API URL if not set.
+        """
+        logging.getLogger("httpx").setLevel(log_level)
+        logging.getLogger("httpcore").setLevel(log_level)
+        super().__init__(
+            headers={"Api-Key": api_key or os.environ.get("ORCA_API_KEY", "")} | dict(Headers(headers or {}).items()),
+            http1=http1,
+            http2=http2,
+            proxy=proxy,
+            timeout=timeout or Timeout(connect=3, read=20, write=10, pool=5),
+            follow_redirects=True,
+            limits=limits or Limits(max_connections=100, max_keepalive_connections=20),
+            max_redirects=max_redirects,
+            event_hooks=event_hooks
+            or {"request": [self._instrument_request], "response": [self._raise_error_for_response]},
+            base_url=base_url or os.environ.get("ORCA_API_URL", "https://api.orcadb.ai/"),
+            transport=transport
+            or RetryTransport(
+                transport=AsyncHTTPTransport(),
+                retry=Retry(
+                    total=5,
+                    backoff_factor=0.5,
+                    allowed_methods=["GET", "POST", "PUT", "PATCH", "DELETE"],
+                    status_forcelist=[429, 500, 502, 503, 504],
+                ),
+            ),
+        )
+
+    @property
+    def api_key(self) -> str:
+        return self.headers["Api-Key"]
+
+    @api_key.setter
+    def api_key(self, api_key: str) -> None:
+        self.headers.update(Headers({"Api-Key": api_key}))
+
+    client_ctx = ContextVar[Self | None]("orca_async_client", default=None)
+    default_client: Self | None = None
+
+    @contextmanager
+    def use(self) -> Generator[None, None, None]:
+        """Context manager to inject this async client into OrcaSDK async methods"""
+        token = self.client_ctx.set(self)
+        try:
+            yield
+        finally:
+            self.client_ctx.reset(token)
+
+    @classmethod
+    def _resolve_client(cls, client: Self | None = None) -> Self:
+        client = client or cls.client_ctx.get() or cls.default_client
+        if not client:
+            client = cls.default_client = cls()
+        return client