@homeflare/distilled-litellm 0.2.0 → 0.4.0

package/src/services/key_management.ts

@@ -1,4 +1,4 @@
-// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.100.0 — see README). Do not edit.
+// AUTO-GENERATED by scripts/generate.ts from .generated-specs (pinned litellm-openapi @ 1.103.0 — see README). Do not edit.
 import * as S from "@distilled.cloud/core/schema";
 import * as Redacted from "effect/Redacted";
 import * as API from "@distilled.cloud/core/api";
@@ -31,6 +31,31 @@ export class Forbidden
     [{ status: 403 }],
   ) {}
 
+/** The caller may not delete a key it named (it is not a proxy admin and not allowed to modify that key). LiteLLM answers 403 with the message `You are not authorized to delete this key`; the key was not deleted. */
+export class KeyDeleteForbidden
+  extends /*@__PURE__*/ T.applyErrorMatchers(
+    /*@__PURE__*/ S.TaggedError<KeyDeleteForbidden>()("KeyDeleteForbidden", {
+      code: S.Number,
+      message: S.String,
+    }).pipe(C.withAuthError),
+    [
+      {
+        status: 403,
+        message: { includes: "not authorized to delete this key" },
+      },
+    ],
+  ) {}
+
+/** No key matches what /key/delete was asked to delete: the alias (or key) is absent, or another caller deleted it first. LiteLLM answers 404 with the message `No keys found`. Whether the key is still there is a question for /key/list, not for this error. */
+export class KeyNotFound
+  extends /*@__PURE__*/ T.applyErrorMatchers(
+    /*@__PURE__*/ S.TaggedError<KeyNotFound>()("KeyNotFound", {
+      code: S.Number,
+      message: S.String,
+    }).pipe(C.withBadRequestError),
+    [{ status: 404, message: { includes: "No keys found" } }],
+  ) {}
+
 export class NotFound
   extends /*@__PURE__*/ T.applyErrorMatchers(
     /*@__PURE__*/ S.TaggedError<NotFound>()("NotFound", {
@@ -49,16 +74,138 @@ export class UnprocessableEntity
     [{ status: 422 }],
   ) {}
 
+export type LiteLLMObjectPermissionBaseAgentAccessGroupsList = Array<string>;
+export const LiteLLMObjectPermissionBaseAgentAccessGroupsList =
+  /*@__PURE__*/ S.Array(
+    S.String,
+  ) as any as S.Schema<LiteLLMObjectPermissionBaseAgentAccessGroupsList>;
+
+export type LiteLLMObjectPermissionBaseAgentsList = Array<string>;
+export const LiteLLMObjectPermissionBaseAgentsList = /*@__PURE__*/ S.Array(
+  S.String,
+) as any as S.Schema<LiteLLMObjectPermissionBaseAgentsList>;
+
+export type LiteLLMObjectPermissionBaseBlockedToolsList = Array<string>;
+export const LiteLLMObjectPermissionBaseBlockedToolsList =
+  /*@__PURE__*/ S.Array(
+    S.String,
+  ) as any as S.Schema<LiteLLMObjectPermissionBaseBlockedToolsList>;
+
+export type LiteLLMObjectPermissionBaseMcpAccessGroupsList = Array<string>;
+export const LiteLLMObjectPermissionBaseMcpAccessGroupsList =
+  /*@__PURE__*/ S.Array(
+    S.String,
+  ) as any as S.Schema<LiteLLMObjectPermissionBaseMcpAccessGroupsList>;
+
+export type LiteLLMObjectPermissionBaseMcpServersList = Array<string>;
+export const LiteLLMObjectPermissionBaseMcpServersList = /*@__PURE__*/ S.Array(
+  S.String,
+) as any as S.Schema<LiteLLMObjectPermissionBaseMcpServersList>;
+
+export type LiteLLMObjectPermissionBaseMcpToolPermissionsValueList =
+  Array<string>;
+export const LiteLLMObjectPermissionBaseMcpToolPermissionsValueList =
+  /*@__PURE__*/ S.Array(
+    S.String,
+  ) as any as S.Schema<LiteLLMObjectPermissionBaseMcpToolPermissionsValueList>;
+
+export type LiteLLMObjectPermissionBaseMcpToolPermissionsMap = {
+  [key: string]:
+    | LiteLLMObjectPermissionBaseMcpToolPermissionsValueList
+    | undefined;
+};
+export const LiteLLMObjectPermissionBaseMcpToolPermissionsMap =
+  /*@__PURE__*/ S.Record(
+    S.String,
+    LiteLLMObjectPermissionBaseMcpToolPermissionsValueList,
+  ) as any as S.Schema<LiteLLMObjectPermissionBaseMcpToolPermissionsMap>;
+
+export type LiteLLMObjectPermissionBaseMcpToolsetsList = Array<string>;
+export const LiteLLMObjectPermissionBaseMcpToolsetsList = /*@__PURE__*/ S.Array(
+  S.String,
+) as any as S.Schema<LiteLLMObjectPermissionBaseMcpToolsetsList>;
+
+export type LiteLLMObjectPermissionBaseModelsList = Array<string>;
+export const LiteLLMObjectPermissionBaseModelsList = /*@__PURE__*/ S.Array(
+  S.String,
+) as any as S.Schema<LiteLLMObjectPermissionBaseModelsList>;
+
+export type LiteLLMObjectPermissionBaseSearchToolsList = Array<string>;
+export const LiteLLMObjectPermissionBaseSearchToolsList = /*@__PURE__*/ S.Array(
+  S.String,
+) as any as S.Schema<LiteLLMObjectPermissionBaseSearchToolsList>;
+
+export type LiteLLMObjectPermissionBaseSkillsList = Array<string>;
+export const LiteLLMObjectPermissionBaseSkillsList = /*@__PURE__*/ S.Array(
+  S.String,
+) as any as S.Schema<LiteLLMObjectPermissionBaseSkillsList>;
+
+export type LiteLLMObjectPermissionBaseVectorStoresList = Array<string>;
+export const LiteLLMObjectPermissionBaseVectorStoresList =
+  /*@__PURE__*/ S.Array(
+    S.String,
+  ) as any as S.Schema<LiteLLMObjectPermissionBaseVectorStoresList>;
+
+export interface LiteLLMObjectPermissionBase {
+  agent_access_groups?: LiteLLMObjectPermissionBaseAgentAccessGroupsList | null;
+  agents?: LiteLLMObjectPermissionBaseAgentsList | null;
+  blocked_tools?: LiteLLMObjectPermissionBaseBlockedToolsList | null;
+  mcp_access_groups?: LiteLLMObjectPermissionBaseMcpAccessGroupsList | null;
+  mcp_servers?: LiteLLMObjectPermissionBaseMcpServersList | null;
+  mcp_tool_permissions?: LiteLLMObjectPermissionBaseMcpToolPermissionsMap | null;
+  mcp_tool_search_enabled?: boolean | null;
+  mcp_toolsets?: LiteLLMObjectPermissionBaseMcpToolsetsList | null;
+  models?: LiteLLMObjectPermissionBaseModelsList | null;
+  search_tools?: LiteLLMObjectPermissionBaseSearchToolsList | null;
+  skills?: LiteLLMObjectPermissionBaseSkillsList | null;
+  vector_stores?: LiteLLMObjectPermissionBaseVectorStoresList | null;
+}
+export const LiteLLMObjectPermissionBase = /*@__PURE__*/ S.suspend(() =>
+  S.Struct({
+    agent_access_groups: S.optional(
+      S.NullOr(LiteLLMObjectPermissionBaseAgentAccessGroupsList),
+    ),
+    agents: S.optional(S.NullOr(LiteLLMObjectPermissionBaseAgentsList)),
+    blocked_tools: S.optional(
+      S.NullOr(LiteLLMObjectPermissionBaseBlockedToolsList),
+    ),
+    mcp_access_groups: S.optional(
+      S.NullOr(LiteLLMObjectPermissionBaseMcpAccessGroupsList),
+    ),
+    mcp_servers: S.optional(
+      S.NullOr(LiteLLMObjectPermissionBaseMcpServersList),
+    ),
+    mcp_tool_permissions: S.optional(
+      S.NullOr(LiteLLMObjectPermissionBaseMcpToolPermissionsMap),
+    ),
+    mcp_tool_search_enabled: S.optional(S.NullOr(S.Boolean)),
+    mcp_toolsets: S.optional(
+      S.NullOr(LiteLLMObjectPermissionBaseMcpToolsetsList),
+    ),
+    models: S.optional(S.NullOr(LiteLLMObjectPermissionBaseModelsList)),
+    search_tools: S.optional(
+      S.NullOr(LiteLLMObjectPermissionBaseSearchToolsList),
+    ),
+    skills: S.optional(S.NullOr(LiteLLMObjectPermissionBaseSkillsList)),
+    vector_stores: S.optional(
+      S.NullOr(LiteLLMObjectPermissionBaseVectorStoresList),
+    ),
+  }),
+).annotate({
+  identifier: "LiteLLMObjectPermissionBase",
+}) as any as S.Schema<LiteLLMObjectPermissionBase>;
+
 export type BulkUpdateKeyRequestItemTagsList = Array<string>;
 export const BulkUpdateKeyRequestItemTagsList = /*@__PURE__*/ S.Array(
   S.String,
 ) as any as S.Schema<BulkUpdateKeyRequestItemTagsList>;
 
-/** Individual key update request item */
+/** One /key/bulk_update item; only the fields it carries are written. */
 export interface BulkUpdateKeyRequestItem {
   budget_id?: string | null;
   key: string;
   max_budget?: number | null;
+  object_permission?: LiteLLMObjectPermissionBase | null;
   tags?: BulkUpdateKeyRequestItemTagsList | null;
   team_id?: string | null;
 }
@@ -67,6 +214,7 @@ export const BulkUpdateKeyRequestItem = /*@__PURE__*/ S.suspend(() =>
     budget_id: S.optional(S.NullOr(S.String)),
     key: S.String,
     max_budget: S.optional(S.NullOr(S.Number)),
+    object_permission: S.optional(S.NullOr(LiteLLMObjectPermissionBase)),
     tags: S.optional(S.NullOr(BulkUpdateKeyRequestItemTagsList)),
     team_id: S.optional(S.NullOr(S.String)),
   }),
@@ -520,120 +668,6 @@ export const GenerateKeyFnKeyGeneratePostRequestModelsList =
     S.Unknown,
   ) as any as S.Schema<GenerateKeyFnKeyGeneratePostRequestModelsList>;
 
-export type LiteLLMObjectPermissionBaseAgentAccessGroupsList = Array<string>;
-export const LiteLLMObjectPermissionBaseAgentAccessGroupsList =
-  /*@__PURE__*/ S.Array(
-    S.String,
-  ) as any as S.Schema<LiteLLMObjectPermissionBaseAgentAccessGroupsList>;
-
-export type LiteLLMObjectPermissionBaseAgentsList = Array<string>;
-export const LiteLLMObjectPermissionBaseAgentsList = /*@__PURE__*/ S.Array(
-  S.String,
-) as any as S.Schema<LiteLLMObjectPermissionBaseAgentsList>;
-
-export type LiteLLMObjectPermissionBaseBlockedToolsList = Array<string>;
-export const LiteLLMObjectPermissionBaseBlockedToolsList =
-  /*@__PURE__*/ S.Array(
-    S.String,
-  ) as any as S.Schema<LiteLLMObjectPermissionBaseBlockedToolsList>;
-
-export type LiteLLMObjectPermissionBaseMcpAccessGroupsList = Array<string>;
-export const LiteLLMObjectPermissionBaseMcpAccessGroupsList =
-  /*@__PURE__*/ S.Array(
-    S.String,
-  ) as any as S.Schema<LiteLLMObjectPermissionBaseMcpAccessGroupsList>;
-
-export type LiteLLMObjectPermissionBaseMcpServersList = Array<string>;
-export const LiteLLMObjectPermissionBaseMcpServersList = /*@__PURE__*/ S.Array(
-  S.String,
-) as any as S.Schema<LiteLLMObjectPermissionBaseMcpServersList>;
-
-export type LiteLLMObjectPermissionBaseMcpToolPermissionsValueList =
-  Array<string>;
-export const LiteLLMObjectPermissionBaseMcpToolPermissionsValueList =
-  /*@__PURE__*/ S.Array(
-    S.String,
-  ) as any as S.Schema<LiteLLMObjectPermissionBaseMcpToolPermissionsValueList>;
-
-export type LiteLLMObjectPermissionBaseMcpToolPermissionsMap = {
-  [key: string]:
-    | LiteLLMObjectPermissionBaseMcpToolPermissionsValueList
-    | undefined;
-};
-export const LiteLLMObjectPermissionBaseMcpToolPermissionsMap =
-  /*@__PURE__*/ S.Record(
-    S.String,
-    LiteLLMObjectPermissionBaseMcpToolPermissionsValueList,
-  ) as any as S.Schema<LiteLLMObjectPermissionBaseMcpToolPermissionsMap>;
-
-export type LiteLLMObjectPermissionBaseMcpToolsetsList = Array<string>;
-export const LiteLLMObjectPermissionBaseMcpToolsetsList = /*@__PURE__*/ S.Array(
-  S.String,
-) as any as S.Schema<LiteLLMObjectPermissionBaseMcpToolsetsList>;
-
-export type LiteLLMObjectPermissionBaseModelsList = Array<string>;
-export const LiteLLMObjectPermissionBaseModelsList = /*@__PURE__*/ S.Array(
-  S.String,
-) as any as S.Schema<LiteLLMObjectPermissionBaseModelsList>;
-
-export type LiteLLMObjectPermissionBaseSearchToolsList = Array<string>;
-export const LiteLLMObjectPermissionBaseSearchToolsList = /*@__PURE__*/ S.Array(
-  S.String,
-) as any as S.Schema<LiteLLMObjectPermissionBaseSearchToolsList>;
-
-export type LiteLLMObjectPermissionBaseVectorStoresList = Array<string>;
-export const LiteLLMObjectPermissionBaseVectorStoresList =
-  /*@__PURE__*/ S.Array(
-    S.String,
-  ) as any as S.Schema<LiteLLMObjectPermissionBaseVectorStoresList>;
-
-export interface LiteLLMObjectPermissionBase {
-  agent_access_groups?: LiteLLMObjectPermissionBaseAgentAccessGroupsList | null;
-  agents?: LiteLLMObjectPermissionBaseAgentsList | null;
-  blocked_tools?: LiteLLMObjectPermissionBaseBlockedToolsList | null;
-  mcp_access_groups?: LiteLLMObjectPermissionBaseMcpAccessGroupsList | null;
-  mcp_servers?: LiteLLMObjectPermissionBaseMcpServersList | null;
-  mcp_tool_permissions?: LiteLLMObjectPermissionBaseMcpToolPermissionsMap | null;
-  mcp_tool_search_enabled?: boolean | null;
-  mcp_toolsets?: LiteLLMObjectPermissionBaseMcpToolsetsList | null;
-  models?: LiteLLMObjectPermissionBaseModelsList | null;
-  search_tools?: LiteLLMObjectPermissionBaseSearchToolsList | null;
-  vector_stores?: LiteLLMObjectPermissionBaseVectorStoresList | null;
-}
-export const LiteLLMObjectPermissionBase = /*@__PURE__*/ S.suspend(() =>
-  S.Struct({
-    agent_access_groups: S.optional(
-      S.NullOr(LiteLLMObjectPermissionBaseAgentAccessGroupsList),
-    ),
-    agents: S.optional(S.NullOr(LiteLLMObjectPermissionBaseAgentsList)),
-    blocked_tools: S.optional(
-      S.NullOr(LiteLLMObjectPermissionBaseBlockedToolsList),
-    ),
-    mcp_access_groups: S.optional(
-      S.NullOr(LiteLLMObjectPermissionBaseMcpAccessGroupsList),
-    ),
-    mcp_servers: S.optional(
-      S.NullOr(LiteLLMObjectPermissionBaseMcpServersList),
-    ),
-    mcp_tool_permissions: S.optional(
-      S.NullOr(LiteLLMObjectPermissionBaseMcpToolPermissionsMap),
-    ),
-    mcp_tool_search_enabled: S.optional(S.NullOr(S.Boolean)),
-    mcp_toolsets: S.optional(
-      S.NullOr(LiteLLMObjectPermissionBaseMcpToolsetsList),
-    ),
-    models: S.optional(S.NullOr(LiteLLMObjectPermissionBaseModelsList)),
-    search_tools: S.optional(
-      S.NullOr(LiteLLMObjectPermissionBaseSearchToolsList),
-    ),
-    vector_stores: S.optional(
-      S.NullOr(LiteLLMObjectPermissionBaseVectorStoresList),
-    ),
-  }),
-).annotate({
-  identifier: "LiteLLMObjectPermissionBase",
-}) as any as S.Schema<LiteLLMObjectPermissionBase>;
-
 export type GenerateKeyFnKeyGeneratePostRequestPermissionsMap = {
   [key: string]: unknown | undefined;
 };
@@ -730,8 +764,10 @@ export interface RetryPolicy {
   AuthenticationErrorRetries?: number | null;
   BadRequestErrorRetries?: number | null;
   ContentPolicyViolationErrorRetries?: number | null;
+  DefaultRetries?: number | null;
   InternalServerErrorRetries?: number | null;
   RateLimitErrorRetries?: number | null;
+  ServiceUnavailableErrorRetries?: number | null;
   TimeoutErrorRetries?: number | null;
 }
 export const RetryPolicy = /*@__PURE__*/ S.suspend(() =>
@@ -739,8 +775,10 @@ export const RetryPolicy = /*@__PURE__*/ S.suspend(() =>
     AuthenticationErrorRetries: S.optional(S.NullOr(S.Number)),
     BadRequestErrorRetries: S.optional(S.NullOr(S.Number)),
     ContentPolicyViolationErrorRetries: S.optional(S.NullOr(S.Number)),
+    DefaultRetries: S.optional(S.NullOr(S.Number)),
     InternalServerErrorRetries: S.optional(S.NullOr(S.Number)),
     RateLimitErrorRetries: S.optional(S.NullOr(S.Number)),
+    ServiceUnavailableErrorRetries: S.optional(S.NullOr(S.Number)),
     TimeoutErrorRetries: S.optional(S.NullOr(S.Number)),
   }),
 ).annotate({ identifier: "RetryPolicy" }) as any as S.Schema<RetryPolicy>;
@@ -754,6 +792,25 @@ export const UpdateRouterConfigModelGroupRetryPolicyMap =
     RetryPolicy,
   ) as any as S.Schema<UpdateRouterConfigModelGroupRetryPolicyMap>;
 
+export type UpdateRouterConfigOptionalPreCallChecksItem =
+  | "prompt_caching"
+  | "router_budget_limiting"
+  | "responses_api_deployment_check"
+  | "deployment_affinity"
+  | "session_affinity"
+  | "forward_client_headers_by_model_group"
+  | "enforce_model_rate_limits"
+  | "encrypted_content_affinity";
+export const UpdateRouterConfigOptionalPreCallChecksItem = S.String;
+
+export type UpdateRouterConfigOptionalPreCallChecksList = Array<
+  UpdateRouterConfigOptionalPreCallChecksItem | (string & {})
+>;
+export const UpdateRouterConfigOptionalPreCallChecksList =
+  /*@__PURE__*/ S.Array(
+    UpdateRouterConfigOptionalPreCallChecksItem,
+  ) as any as S.Schema<UpdateRouterConfigOptionalPreCallChecksList>;
+
 export type RoutingGroupModelsList = Array<string>;
 export const RoutingGroupModelsList = /*@__PURE__*/ S.Array(
   S.String,
@@ -810,6 +867,7 @@ export interface UpdateRouterConfig {
   model_group_alias?: UpdateRouterConfigModelGroupAliasMap | null;
   model_group_retry_policy?: UpdateRouterConfigModelGroupRetryPolicyMap | null;
   num_retries?: number | null;
+  optional_pre_call_checks?: UpdateRouterConfigOptionalPreCallChecksList | null;
   retry_after?: number | null;
   retry_policy?: RetryPolicy | null;
   routing_groups?: UpdateRouterConfigRoutingGroupsList | null;
@@ -817,6 +875,7 @@ export interface UpdateRouterConfig {
   routing_strategy_args?: UpdateRouterConfigRoutingStrategyArgsMap | null;
   tag_routing_prefix?: string | null;
   timeout?: number | null;
+  weights?: unknown | null;
 }
 export const UpdateRouterConfig = /*@__PURE__*/ S.suspend(() =>
   S.Struct({
@@ -838,6 +897,9 @@ export const UpdateRouterConfig = /*@__PURE__*/ S.suspend(() =>
       S.NullOr(UpdateRouterConfigModelGroupRetryPolicyMap),
     ),
     num_retries: S.optional(S.NullOr(S.Number)),
+    optional_pre_call_checks: S.optional(
+      S.NullOr(UpdateRouterConfigOptionalPreCallChecksList),
+    ),
     retry_after: S.optional(S.NullOr(S.Number)),
     retry_policy: S.optional(S.NullOr(RetryPolicy)),
     routing_groups: S.optional(S.NullOr(UpdateRouterConfigRoutingGroupsList)),
@@ -847,6 +909,7 @@ export const UpdateRouterConfig = /*@__PURE__*/ S.suspend(() =>
     ),
     tag_routing_prefix: S.optional(S.NullOr(S.String)),
     timeout: S.optional(S.NullOr(S.Number)),
+    weights: S.optional(S.NullOr(S.Unknown)),
   }),
 ).annotate({
   identifier: "UpdateRouterConfig",
@@ -900,6 +963,7 @@ export interface GenerateKeyFnKeyGeneratePostRequest {
   disable_global_guardrails?: boolean | null;
   duration?: string | null;
   enable_prompt_caching?: boolean | null;
+  end_user_budget_id?: string | null;
   enforced_params?: GenerateKeyFnKeyGeneratePostRequestEnforcedParamsList | null;
   guardrails?: GenerateKeyFnKeyGeneratePostRequestGuardrailsList | null;
   key?: string | null;
@@ -935,6 +999,7 @@ export interface GenerateKeyFnKeyGeneratePostRequest {
   tags?: GenerateKeyFnKeyGeneratePostRequestTagsList | null;
   team_id?: string | null;
   throttle_on_budget_exceeded?: boolean | null;
+  tpd_limit?: number | null;
   tpm_limit?: number | null;
   tpm_limit_type?:
     | GenerateKeyFnKeyGeneratePostRequestTpmLimitType
@@ -985,6 +1050,7 @@ export const GenerateKeyFnKeyGeneratePostRequest = /*@__PURE__*/ S.suspend(() =>
     disable_global_guardrails: S.optional(S.NullOr(S.Boolean)),
     duration: S.optional(S.NullOr(S.String)),
     enable_prompt_caching: S.optional(S.NullOr(S.Boolean)),
+    end_user_budget_id: S.optional(S.NullOr(S.String)),
     enforced_params: S.optional(
       S.NullOr(GenerateKeyFnKeyGeneratePostRequestEnforcedParamsList),
     ),
@@ -1039,6 +1105,7 @@ export const GenerateKeyFnKeyGeneratePostRequest = /*@__PURE__*/ S.suspend(() =>
     tags: S.optional(S.NullOr(GenerateKeyFnKeyGeneratePostRequestTagsList)),
     team_id: S.optional(S.NullOr(S.String)),
     throttle_on_budget_exceeded: S.optional(S.NullOr(S.Boolean)),
+    tpd_limit: S.optional(S.NullOr(S.Number)),
     tpm_limit: S.optional(S.NullOr(S.Number)),
     tpm_limit_type: S.optional(
       S.NullOr(GenerateKeyFnKeyGeneratePostRequestTpmLimitType),
@@ -1241,6 +1308,7 @@ export interface GenerateKeyResponse {
   disable_global_guardrails?: boolean | null;
   duration?: string | null;
   enable_prompt_caching?: boolean | null;
+  end_user_budget_id?: string | null;
   enforced_params?: GenerateKeyResponseEnforcedParamsList | null;
   expires?: string | null;
   guardrails?: GenerateKeyResponseGuardrailsList | null;
@@ -1273,6 +1341,7 @@ export interface GenerateKeyResponse {
   throttle_on_budget_exceeded?: boolean | null;
   token?: string | null;
   token_id?: string | null;
+  tpd_limit?: number | null;
   tpm_limit?: number | null;
   tpm_limit_type?: GenerateKeyResponseTpmLimitType | null;
   updated_at?: string | null;
@@ -1313,6 +1382,7 @@ export const GenerateKeyResponse = /*@__PURE__*/ S.suspend(() =>
     disable_global_guardrails: S.optional(S.NullOr(S.Boolean)),
     duration: S.optional(S.NullOr(S.String)),
     enable_prompt_caching: S.optional(S.NullOr(S.Boolean)),
+    end_user_budget_id: S.optional(S.NullOr(S.String)),
     enforced_params: S.optional(
       S.NullOr(GenerateKeyResponseEnforcedParamsList),
     ),
@@ -1349,6 +1419,7 @@ export const GenerateKeyResponse = /*@__PURE__*/ S.suspend(() =>
     throttle_on_budget_exceeded: S.optional(S.NullOr(S.Boolean)),
     token: S.optional(S.NullOr(S.String)),
     token_id: S.optional(S.NullOr(S.String)),
+    tpd_limit: S.optional(S.NullOr(S.Number)),
     tpm_limit: S.optional(S.NullOr(S.Number)),
     tpm_limit_type: S.optional(S.NullOr(GenerateKeyResponseTpmLimitType)),
     updated_at: S.optional(S.NullOr(S.String)),
@@ -1577,6 +1648,7 @@ export interface GenerateServiceAccountKeyFnKeyServiceAccountGeneratePostRequest
   disable_global_guardrails?: boolean | null;
   duration?: string | null;
   enable_prompt_caching?: boolean | null;
+  end_user_budget_id?: string | null;
   enforced_params?: GenerateServiceAccountKeyFnKeyServiceAccountGeneratePostRequestEnforcedParamsList | null;
   guardrails?: GenerateServiceAccountKeyFnKeyServiceAccountGeneratePostRequestGuardrailsList | null;
   key?: string | null;
@@ -1612,6 +1684,7 @@ export interface GenerateServiceAccountKeyFnKeyServiceAccountGeneratePostRequest
   tags?: GenerateServiceAccountKeyFnKeyServiceAccountGeneratePostRequestTagsList | null;
   team_id?: string | null;
   throttle_on_budget_exceeded?: boolean | null;
+  tpd_limit?: number | null;
   tpm_limit?: number | null;
   tpm_limit_type?:
     | GenerateServiceAccountKeyFnKeyServiceAccountGeneratePostRequestTpmLimitType
@@ -1681,6 +1754,7 @@ export const GenerateServiceAccountKeyFnKeyServiceAccountGeneratePostRequest =
       disable_global_guardrails: S.optional(S.NullOr(S.Boolean)),
       duration: S.optional(S.NullOr(S.String)),
       enable_prompt_caching: S.optional(S.NullOr(S.Boolean)),
+      end_user_budget_id: S.optional(S.NullOr(S.String)),
       enforced_params: S.optional(
         S.NullOr(
           GenerateServiceAccountKeyFnKeyServiceAccountGeneratePostRequestEnforcedParamsList,
@@ -1767,6 +1841,7 @@ export const GenerateServiceAccountKeyFnKeyServiceAccountGeneratePostRequest =
       ),
       team_id: S.optional(S.NullOr(S.String)),
       throttle_on_budget_exceeded: S.optional(S.NullOr(S.Boolean)),
+      tpd_limit: S.optional(S.NullOr(S.Number)),
       tpm_limit: S.optional(S.NullOr(S.Number)),
       tpm_limit_type: S.optional(
         S.NullOr(
@@ -1800,7 +1875,7 @@ export const GenerateServiceAccountKeyFnKeyServiceAccountGeneratePostResponse =
   }) as any as S.Schema<GenerateServiceAccountKeyFnKeyServiceAccountGeneratePostResponse>;
 
 export interface GetInfoKeyFnKeyInfoRequest {
-  /** Key in the request parameters */
+  /** Key to look up. Pass the key's sha256 hash so the raw key stays out of URLs and access logs. Example key='d5345c0ecc68ae6295c69f91926b2bd379e25481a40c34b5884d157a9f65d8fa' */
   key?: string;
 }
 export const GetInfoKeyFnKeyInfoRequest = /*@__PURE__*/ S.suspend(() =>
@@ -1880,8 +1955,10 @@ export interface ListKeysKeyListGetRequest {
   organization_id?: string;
   /** Filter keys by key hash */
   key_hash?: string;
-  /** Filter keys by key alias. Exact match by default; set substring_matching=true (admin only) for case-insensitive substring matching. */
+  /** Filter keys by key alias. Exact match by default; set substring_matching=true for case-insensitive substring matching. */
   key_alias?: string;
+  /** Combined search: matches keys whose token (key hash) equals the value OR whose key_alias contains it (case-insensitive). */
+  search?: string;
   /** Return full key object */
   return_full_object?: boolean;
   /** Include all keys for teams that user is an admin of. */
@@ -1894,7 +1971,7 @@ export interface ListKeysKeyListGetRequest {
   sort_order?: string;
   /** Expand related objects (e.g. 'user') */
   expand?: ListKeysKeyListGetRequestExpandList;
-  /** Filter by status (e.g. 'deleted') */
+  /** Filter by status: 'active' (not blocked, not expired), 'expired' (not blocked, past expiry), 'revoked' (blocked) or 'deleted' (archived keys). Omit to return live keys regardless of status. */
   status?: string;
   /** Filter keys by project ID */
   project_id?: string;
@@ -1902,7 +1979,7 @@ export interface ListKeysKeyListGetRequest {
   access_group_id?: string;
   /** Filter keys by agent ID */
   agent_id?: string;
-  /** If true (proxy admins only), match user_id/key_alias as case-insensitive substrings instead of exact values. Defaults to false: /key/list matched these exactly before substring search was added, and an exact user_id/key_alias filter must never return another user's keys. */
+  /** If true, match key_alias (any caller) and user_id (proxy admins only) as case-insensitive substrings instead of exact values. Defaults to false: /key/list matched these exactly before substring search was added, and an exact user_id filter must never return another user's keys. */
   substring_matching?: boolean;
   /** Filter keys by expiration. 'expired' returns keys whose expires is in the past; 'active' returns keys that never expire or expire in the future. Omit to return keys regardless of expiration. */
   expires?: string;
@@ -1916,6 +1993,7 @@ export const ListKeysKeyListGetRequest = /*@__PURE__*/ S.suspend(() =>
     organization_id: S.optional(S.String.pipe(T.Query())),
     key_hash: S.optional(S.String.pipe(T.Query())),
     key_alias: S.optional(S.String.pipe(T.Query())),
+    search: S.optional(S.String.pipe(T.Query())),
     return_full_object: S.optional(S.Boolean.pipe(T.Query())),
     include_team_keys: S.optional(S.Boolean.pipe(T.Query())),
     include_created_by_keys: S.optional(S.Boolean.pipe(T.Query())),
@@ -2061,6 +2139,11 @@ export const LiteLLMObjectPermissionTableSearchToolsList =
     S.String,
   ) as any as S.Schema<LiteLLMObjectPermissionTableSearchToolsList>;
 
+export type LiteLLMObjectPermissionTableSkillsList = Array<string>;
+export const LiteLLMObjectPermissionTableSkillsList = /*@__PURE__*/ S.Array(
+  S.String,
+) as any as S.Schema<LiteLLMObjectPermissionTableSkillsList>;
+
 export type LiteLLMObjectPermissionTableVectorStoresList = Array<string>;
 export const LiteLLMObjectPermissionTableVectorStoresList =
   /*@__PURE__*/ S.Array(
@@ -2080,6 +2163,7 @@ export interface LiteLLMObjectPermissionTable {
   models?: LiteLLMObjectPermissionTableModelsList | null;
   object_permission_id: string;
   search_tools?: LiteLLMObjectPermissionTableSearchToolsList | null;
+  skills?: LiteLLMObjectPermissionTableSkillsList | null;
   vector_stores?: LiteLLMObjectPermissionTableVectorStoresList | null;
 }
 export const LiteLLMObjectPermissionTable = /*@__PURE__*/ S.suspend(() =>
@@ -2109,6 +2193,7 @@ export const LiteLLMObjectPermissionTable = /*@__PURE__*/ S.suspend(() =>
     search_tools: S.optional(
       S.NullOr(LiteLLMObjectPermissionTableSearchToolsList),
     ),
+    skills: S.optional(S.NullOr(LiteLLMObjectPermissionTableSkillsList)),
     vector_stores: S.optional(
       S.NullOr(LiteLLMObjectPermissionTableVectorStoresList),
     ),
@@ -2234,6 +2319,14 @@ export const UserAPIKeyAuthTeamModelAliasesMap = /*@__PURE__*/ S.Record(
   S.Unknown,
 ) as any as S.Schema<UserAPIKeyAuthTeamModelAliasesMap>;
 
+export type UserAPIKeyAuthTeamModelMaxBudgetMap = {
+  [key: string]: unknown | undefined;
+};
+export const UserAPIKeyAuthTeamModelMaxBudgetMap = /*@__PURE__*/ S.Record(
+  S.String,
+  S.Unknown,
+) as any as S.Schema<UserAPIKeyAuthTeamModelMaxBudgetMap>;
+
 export type UserAPIKeyAuthTeamModelsList = Array<unknown>;
 export const UserAPIKeyAuthTeamModelsList = /*@__PURE__*/ S.Array(
   S.Unknown,
@@ -2291,6 +2384,7 @@ export interface UserAPIKeyAuth {
   end_user_model_max_budget?: UserAPIKeyAuthEndUserModelMaxBudgetMap | null;
   end_user_object_permission?: LiteLLMObjectPermissionTable | null;
   end_user_rpm_limit?: number | null;
+  end_user_tpd_limit?: number | null;
   end_user_tpm_limit?: number | null;
   expires?: string | null;
   is_session_token?: boolean;
@@ -2342,14 +2436,18 @@ export interface UserAPIKeyAuth {
   team_member_tpm_limit?: number | null;
   team_metadata?: UserAPIKeyAuthTeamMetadataMap | null;
   team_model_aliases?: UserAPIKeyAuthTeamModelAliasesMap | null;
+  team_model_max_budget?: UserAPIKeyAuthTeamModelMaxBudgetMap | null;
   team_models?: UserAPIKeyAuthTeamModelsList;
   team_object_permission?: LiteLLMObjectPermissionTable | null;
   team_object_permission_id?: string | null;
   team_rpm_limit?: number | null;
   team_soft_budget?: number | null;
   team_spend?: number | null;
+  team_tpd_limit?: number | null;
   team_tpm_limit?: number | null;
   token?: string | null;
+  total_spend?: number;
+  tpd_limit?: number | null;
   tpm_limit?: number | null;
   tpm_limit_per_model?: UserAPIKeyAuthTpmLimitPerModelMap | null;
   updated_at?: string | null;
@@ -2397,6 +2495,7 @@ export const UserAPIKeyAuth = /*@__PURE__*/ S.suspend(() =>
       S.NullOr(LiteLLMObjectPermissionTable),
     ),
     end_user_rpm_limit: S.optional(S.NullOr(S.Number)),
+    end_user_tpd_limit: S.optional(S.NullOr(S.Number)),
     end_user_tpm_limit: S.optional(S.NullOr(S.Number)),
     expires: S.optional(S.NullOr(S.String)),
     is_session_token: S.optional(S.Boolean),
@@ -2454,14 +2553,20 @@ export const UserAPIKeyAuth = /*@__PURE__*/ S.suspend(() =>
     team_member_tpm_limit: S.optional(S.NullOr(S.Number)),
     team_metadata: S.optional(S.NullOr(UserAPIKeyAuthTeamMetadataMap)),
     team_model_aliases: S.optional(S.NullOr(UserAPIKeyAuthTeamModelAliasesMap)),
+    team_model_max_budget: S.optional(
+      S.NullOr(UserAPIKeyAuthTeamModelMaxBudgetMap),
+    ),
     team_models: S.optional(UserAPIKeyAuthTeamModelsList),
     team_object_permission: S.optional(S.NullOr(LiteLLMObjectPermissionTable)),
     team_object_permission_id: S.optional(S.NullOr(S.String)),
     team_rpm_limit: S.optional(S.NullOr(S.Number)),
     team_soft_budget: S.optional(S.NullOr(S.Number)),
     team_spend: S.optional(S.NullOr(S.Number)),
+    team_tpd_limit: S.optional(S.NullOr(S.Number)),
     team_tpm_limit: S.optional(S.NullOr(S.Number)),
     token: S.optional(S.NullOr(S.String)),
+    total_spend: S.optional(S.Number),
+    tpd_limit: S.optional(S.NullOr(S.Number)),
     tpm_limit: S.optional(S.NullOr(S.Number)),
     tpm_limit_per_model: S.optional(
       S.NullOr(UserAPIKeyAuthTpmLimitPerModelMap),
@@ -2649,6 +2754,7 @@ export interface LiteLLMDeletedVerificationToken {
   object_permission?: LiteLLMObjectPermissionTable | null;
   object_permission_id?: string | null;
   org_id?: string | null;
+  organization_id?: string | null;
   permissions?: LiteLLMDeletedVerificationTokenPermissionsMap;
   project_id?: string | null;
   rotation_count?: number | null;
@@ -2660,6 +2766,8 @@ export interface LiteLLMDeletedVerificationToken {
   spend?: number;
   team_id?: string | null;
   token?: string | null;
+  total_spend?: number;
+  tpd_limit?: number | null;
   tpm_limit?: number | null;
   updated_at?: string | null;
   updated_by?: string | null;
@@ -2718,6 +2826,7 @@ export const LiteLLMDeletedVerificationToken = /*@__PURE__*/ S.suspend(() =>
     object_permission: S.optional(S.NullOr(LiteLLMObjectPermissionTable)),
     object_permission_id: S.optional(S.NullOr(S.String)),
     org_id: S.optional(S.NullOr(S.String)),
+    organization_id: S.optional(S.NullOr(S.String)),
     permissions: S.optional(LiteLLMDeletedVerificationTokenPermissionsMap),
     project_id: S.optional(S.NullOr(S.String)),
     rotation_count: S.optional(S.NullOr(S.Number)),
@@ -2731,6 +2840,8 @@ export const LiteLLMDeletedVerificationToken = /*@__PURE__*/ S.suspend(() =>
     spend: S.optional(S.Number),
     team_id: S.optional(S.NullOr(S.String)),
     token: S.optional(S.NullOr(S.String)),
+    total_spend: S.optional(S.Number),
+    tpd_limit: S.optional(S.NullOr(S.Number)),
     tpm_limit: S.optional(S.NullOr(S.Number)),
     updated_at: S.optional(S.NullOr(S.String)),
     updated_by: S.optional(S.NullOr(S.String)),
@@ -2941,6 +3052,8 @@ export interface LiteLLMVerificationToken {
   spend?: number;
   team_id?: string | null;
   token?: string | null;
+  total_spend?: number;
+  tpd_limit?: number | null;
   tpm_limit?: number | null;
   updated_at?: string | null;
   updated_by?: string | null;
@@ -3003,6 +3116,8 @@ export const LiteLLMVerificationToken = /*@__PURE__*/ S.suspend(() =>
     spend: S.optional(S.Number),
     team_id: S.optional(S.NullOr(S.String)),
     token: S.optional(S.NullOr(S.String)),
+    total_spend: S.optional(S.Number),
+    tpd_limit: S.optional(S.NullOr(S.Number)),
     tpm_limit: S.optional(S.NullOr(S.Number)),
     updated_at: S.optional(S.NullOr(S.String)),
     updated_by: S.optional(S.NullOr(S.String)),
@@ -3304,6 +3419,7 @@ export interface RegenerateKeyRequest {
   disable_global_guardrails?: boolean | null;
   duration?: string | null;
   enable_prompt_caching?: boolean | null;
+  end_user_budget_id?: string | null;
   enforced_params?: RegenerateKeyRequestEnforcedParamsList | null;
   grace_period?: string | null;
   guardrails?: RegenerateKeyRequestGuardrailsList | null;
@@ -3339,6 +3455,7 @@ export interface RegenerateKeyRequest {
   tags?: RegenerateKeyRequestTagsList | null;
   team_id?: string | null;
   throttle_on_budget_exceeded?: boolean | null;
+  tpd_limit?: number | null;
   tpm_limit?: number | null;
   tpm_limit_type?: RegenerateKeyRequestTpmLimitType | (string & {}) | null;
   user_id?: string | null;
@@ -3376,6 +3493,7 @@ export const RegenerateKeyRequest = /*@__PURE__*/ S.suspend(() =>
     disable_global_guardrails: S.optional(S.NullOr(S.Boolean)),
     duration: S.optional(S.NullOr(S.String)),
     enable_prompt_caching: S.optional(S.NullOr(S.Boolean)),
+    end_user_budget_id: S.optional(S.NullOr(S.String)),
     enforced_params: S.optional(
       S.NullOr(RegenerateKeyRequestEnforcedParamsList),
     ),
@@ -3413,6 +3531,7 @@ export const RegenerateKeyRequest = /*@__PURE__*/ S.suspend(() =>
     tags: S.optional(S.NullOr(RegenerateKeyRequestTagsList)),
     team_id: S.optional(S.NullOr(S.String)),
     throttle_on_budget_exceeded: S.optional(S.NullOr(S.Boolean)),
+    tpd_limit: S.optional(S.NullOr(S.Number)),
     tpm_limit: S.optional(S.NullOr(S.Number)),
     tpm_limit_type: S.optional(S.NullOr(RegenerateKeyRequestTpmLimitType)),
     user_id: S.optional(S.NullOr(S.String)),
@@ -3744,6 +3863,7 @@ export interface UpdateKeyFnKeyUpdatePostRequest {
   disable_global_guardrails?: boolean | null;
   duration?: string | null;
   enable_prompt_caching?: boolean | null;
+  end_user_budget_id?: string | null;
   enforced_params?: UpdateKeyFnKeyUpdatePostRequestEnforcedParamsList | null;
   guardrails?: UpdateKeyFnKeyUpdatePostRequestGuardrailsList | null;
   key?: string | null;
@@ -3760,6 +3880,8 @@ export interface UpdateKeyFnKeyUpdatePostRequest {
   organization_id?: string | null;
   permissions?: UpdateKeyFnKeyUpdatePostRequestPermissionsMap | null;
   policies?: UpdateKeyFnKeyUpdatePostRequestPoliciesList | null;
+  /** Omit to retain the project, or send null to detach. Assigning a different project is not supported. */
+  project_id?: string | null;
   prompts?: UpdateKeyFnKeyUpdatePostRequestPromptsList | null;
   rotation_interval?: string | null;
   router_settings?: UpdateRouterConfig | null;
@@ -3768,6 +3890,7 @@ export interface UpdateKeyFnKeyUpdatePostRequest {
     | UpdateKeyFnKeyUpdatePostRequestRpmLimitType
     | (string & {})
     | null;
+  soft_budget?: number | null;
   spend?: number | null;
   tag_rpm_limit?: UpdateKeyFnKeyUpdatePostRequestTagRpmLimitMap | null;
   tags?: UpdateKeyFnKeyUpdatePostRequestTagsList | null;
@@ -3775,6 +3898,7 @@ export interface UpdateKeyFnKeyUpdatePostRequest {
   temp_budget_expiry?: string | null;
   temp_budget_increase?: number | null;
   throttle_on_budget_exceeded?: boolean | null;
+  tpd_limit?: number | null;
   tpm_limit?: number | null;
   tpm_limit_type?:
     | UpdateKeyFnKeyUpdatePostRequestTpmLimitType
@@ -3821,6 +3945,7 @@ export const UpdateKeyFnKeyUpdatePostRequest = /*@__PURE__*/ S.suspend(() =>
     disable_global_guardrails: S.optional(S.NullOr(S.Boolean)),
     duration: S.optional(S.NullOr(S.String)),
     enable_prompt_caching: S.optional(S.NullOr(S.Boolean)),
+    end_user_budget_id: S.optional(S.NullOr(S.String)),
     enforced_params: S.optional(
       S.NullOr(UpdateKeyFnKeyUpdatePostRequestEnforcedParamsList),
     ),
@@ -3851,6 +3976,7 @@ export const UpdateKeyFnKeyUpdatePostRequest = /*@__PURE__*/ S.suspend(() =>
       S.NullOr(UpdateKeyFnKeyUpdatePostRequestPermissionsMap),
     ),
     policies: S.optional(S.NullOr(UpdateKeyFnKeyUpdatePostRequestPoliciesList)),
+    project_id: S.optional(S.NullOr(S.String)),
     prompts: S.optional(S.NullOr(UpdateKeyFnKeyUpdatePostRequestPromptsList)),
     rotation_interval: S.optional(S.NullOr(S.String)),
     router_settings: S.optional(S.NullOr(UpdateRouterConfig)),
@@ -3858,6 +3984,7 @@ export const UpdateKeyFnKeyUpdatePostRequest = /*@__PURE__*/ S.suspend(() =>
     rpm_limit_type: S.optional(
       S.NullOr(UpdateKeyFnKeyUpdatePostRequestRpmLimitType),
     ),
+    soft_budget: S.optional(S.NullOr(S.Number)),
     spend: S.optional(S.NullOr(S.Number)),
     tag_rpm_limit: S.optional(
       S.NullOr(UpdateKeyFnKeyUpdatePostRequestTagRpmLimitMap),
@@ -3867,6 +3994,7 @@ export const UpdateKeyFnKeyUpdatePostRequest = /*@__PURE__*/ S.suspend(() =>
     temp_budget_expiry: S.optional(S.NullOr(S.String)),
     temp_budget_increase: S.optional(S.NullOr(S.Number)),
     throttle_on_budget_exceeded: S.optional(S.NullOr(S.Boolean)),
+    tpd_limit: S.optional(S.NullOr(S.Number)),
     tpm_limit: S.optional(S.NullOr(S.Number)),
     tpm_limit_type: S.optional(
       S.NullOr(UpdateKeyFnKeyUpdatePostRequestTpmLimitType),
@@ -3893,7 +4021,7 @@ export type BulkUpdateKeysKeyBulkUpdatePostError =
   | Forbidden
   | UnprocessableEntity
   | LitellmOpError;
-/** Bulk Update Keys Bulk update multiple keys at once. This endpoint allows updating multiple keys in a single request. Each key update is processed independently - if some updates fail, others will still succeed. Parameters: - keys: List[BulkUpdateKeyRequestItem] - List of key update requests, each containing: - key: str - The key identifier (token) to update - budget_id: Optional[str] - Budget ID associated with the key - max_budget: Optional[float] - Max budget for key - team_id: Optional[str] - Team ID associated with key - tags: Optional[List[str]] - Tags for organizing keys Returns: - total_requested: int - Total number of keys requested for update - successful_updates: List[SuccessfulKeyUpdate] - List of successfully updated keys with their updated info - failed_updates: List[FailedKeyUpdate] - List of failed updates with key_info and failed_reason Example request: ```bash curl --location 'http://0.0.0.0:4000/key/bulk_update' --header 'Authorization: Bearer sk-1234' --header 'Content-Type: application/json' --data '{ "keys": [ { "key": "sk-1234", "max_budget": 100.0, "team_id": "team-123", "tags": ["production", "api"] }, { "key": "sk-5678", "budget_id": "budget-456", "tags": ["staging"] } ] }' ``` */
+/** Bulk Update Keys Bulk update multiple keys at once. This endpoint allows updating multiple keys in a single request. Each key update is processed independently - if some updates fail, others will still succeed. Parameters: - keys: List[BulkUpdateKeyRequestItem] - List of key update requests, each containing: - key: str - The key identifier (token) to update - budget_id: Optional[str] - Budget ID associated with the key - max_budget: Optional[float] - Max budget for key - team_id: Optional[str] - Team ID associated with key - tags: Optional[List[str]] - Tags for organizing keys - object_permission: Optional[LiteLLM_ObjectPermissionBase] - key-specific object permission, as on /key/update Only the fields an item carries are written: a field left out keeps its current value, and a field sent explicitly, null included, is applied exactly as /key/update applies it. Returns: - total_requested: int - Total number of keys requested for update - successful_updates: List[SuccessfulKeyUpdate] - List of successfully updated keys with their updated info - failed_updates: List[FailedKeyUpdate] - List of failed updates with key_info and failed_reason Example request: ```bash curl --location 'http://0.0.0.0:4000/key/bulk_update' --header 'Authorization: Bearer sk-1234' --header 'Content-Type: application/json' --data '{ "keys": [ { "key": "sk-1234", "max_budget": 100.0, "team_id": "team-123", "tags": ["production", "api"] }, { "key": "sk-5678", "budget_id": "budget-456", "tags": ["staging"] } ] }' ``` */
 export const bulkUpdateKeysKeyBulkUpdatePost: API.OperationMethod<
   BulkUpdateKeysKeyBulkUpdatePostRequest,
   BulkUpdateKeyResponse,
@@ -3927,6 +4055,8 @@ export const bulkUpdateTeamKeysTeamKeyBulkUpdatePost: API.OperationMethod<
 export type DeleteKeyFnKeyDeletePostError =
   | BadRequest
   | UnprocessableEntity
+  | KeyNotFound
+  | KeyDeleteForbidden
   | LitellmOpError;
 /** Delete Key Fn Delete a key from the key management system. Parameters:: - keys (List[str]): A list of keys or hashed keys to delete. Example {"keys": ["sk-QWrxEynunsNpV1zT48HIrw", "837e17519f44683334df5291321d97b8bf1098cd490e49e215f6fea935aa28be"]} - key_aliases (List[str]): A list of key aliases to delete. Can be passed instead of `keys`.Example {"key_aliases": ["alias1", "alias2"]} Returns: - deleted_keys (List[str]): A list of deleted keys. Example {"deleted_keys": ["sk-QWrxEynunsNpV1zT48HIrw", "837e17519f44683334df5291321d97b8bf1098cd490e49e215f6fea935aa28be"]} Example: ```bash curl --location 'http://0.0.0.0:4000/key/delete' --header 'Authorization: Bearer sk-1234' --header 'Content-Type: application/json' --data '{ "keys": ["sk-QWrxEynunsNpV1zT48HIrw"] }' ``` Raises: HTTPException: If an error occurs during key deletion. */
 export const deleteKeyFnKeyDeletePost: API.OperationMethod<
@@ -3937,7 +4067,7 @@ export const deleteKeyFnKeyDeletePost: API.OperationMethod<
 > = /*@__PURE__*/ API.make(() => ({
   input: DeleteKeyFnKeyDeletePostRequest,
   output: DeleteKeyFnKeyDeletePostResponse,
-  errors: [BadRequest, UnprocessableEntity],
+  errors: [BadRequest, UnprocessableEntity, KeyNotFound, KeyDeleteForbidden],
   protocol: LitellmProtocol,
   retry: Retry.Retry,
 }));
@@ -3947,7 +4077,7 @@ export type GenerateKeyFnKeyGeneratePostError =
   | Forbidden
   | UnprocessableEntity
   | LitellmOpError;
-/** Generate Key Fn Generate an API key based on the provided data. Docs: https://docs.litellm.ai/docs/proxy/virtual_keys Parameters: - duration: Optional[str] - Specify the length of time the token is valid for. You can set duration as seconds ("30s"), minutes ("30m"), hours ("30h"), days ("30d"). - key_alias: Optional[str] - User defined key alias - key: Optional[str] - User defined key value. Must start with 'sk-' and be at least 16 characters long. If not set, a 16-digit unique sk-key is created for you. - team_id: Optional[str] - The team id of the key - user_id: Optional[str] - The user id of the key - agent_id: Optional[str] - The agent id associated with the key. - organization_id: Optional[str] - The organization id of the key. If not set, and team_id is set, the organization id will be the same as the team id. If conflict, an error will be raised. - project_id: Optional[str] - The project id of the key. When set, models and max_budget are validated against the project's limits. - budget_id: Optional[str] - The budget id associated with the key. Created by calling `/budget/new`. - models: Optional[list] - Model_name's a user is allowed to call. (if empty, key is allowed to call all models) - aliases: Optional[dict] - Any alias mappings, on top of anything in the config.yaml model list. - https://docs.litellm.ai/docs/proxy/virtual_keys#managing-auth---upgradedowngrade-models - config: Optional[dict] - any key-specific configs, overrides config in config.yaml - spend: Optional[int] - Amount spent by key. Default is 0. Will be updated by proxy whenever key is used. https://docs.litellm.ai/docs/proxy/virtual_keys#managing-auth---tracking-spend - send_invite_email: Optional[bool] - Whether to send an invite email to the user_id, with the generate key - max_budget: Optional[float] - Specify max budget for a given key. - budget_duration: Optional[str] - Budget is reset at the end of specified duration. If not set, budget is never reset. You can set duration as seconds ("30s"), minutes ("30m"), hours ("30h"), days ("30d"). - max_parallel_requests: Optional[int] - Rate limit a user based on the number of parallel requests. Raises 429 error, if user's parallel requests > x. - metadata: Optional[dict] - Metadata for key, store information for key. Example metadata = {"team": "core-infra", "app": "app2", "email": "ishaan@berri.ai" } - guardrails: Optional[List[str]] - List of active guardrails for the key - policies: Optional[List[str]] - List of policy names to apply to the key. Policies define guardrails, conditions, and inheritance rules. - disable_global_guardrails: Optional[bool] - Whether to disable global guardrails for the key. - throttle_on_budget_exceeded: Optional[bool] - When the key exceeds its max_budget, throttle its tpm/rpm to the global budget_exceeded_throttle_percentage instead of blocking the key entirely. - enable_prompt_caching: Optional[bool] - Auto-inject prompt caching breakpoints (Anthropic cache_control markers) on requests made with this key. Anthropic and Bedrock Claude models only. - permissions: Optional[dict] - key-specific permissions. Currently just used for turning off pii masking (if connected). Example - {"pii": false} - model_max_budget: Optional[Dict[str, BudgetConfig]] - Model-specific budgets {"gpt-4": {"budget_limit": 0.0005, "time_period": "30d"}}}. IF null or {} then no model specific budget. - budget_fallbacks: Optional[Dict[str, List[str]]] - Per-model fallback chain tried in order when that model's own `model_max_budget` is exceeded, e.g. {"gpt-4o": ["gpt-4o-mini"]}. - model_rpm_limit: Optional[dict] - key-specific model rpm limit. Example - {"text-davinci-002": 1000, "gpt-3.5-turbo": 1000}. IF null or {} then no model specific rpm limit. - model_tpm_limit: Optional[dict] - key-specific model tpm limit. Example - {"text-davinci-002": 1000, "gpt-3.5-turbo": 1000}. IF null or {} then no model specific tpm limit. - default_estimated_output_tokens: Optional[int] - Proxy admin only. Expected output tokens reserved for TPM limiting when a request omits max_tokens. Positive integer. Falls back to the team setting, then to the built-in estimate. - default_estimated_output_tokens_per_model: Optional[dict] - Proxy admin only. Per-model override of the above. Example - {"gpt-4": 4096, "gpt-3.5-turbo": 1024}. Takes precedence over the key-wide value. - mcp_rpm_limit: Optional[dict] - key-specific per-MCP-server rpm limit, keyed by MCP server name (alias if set, else the configured name). Example - {"github": 100, "slack": 200}. IF null or {} then no MCP-specific rpm limit. - tag_rpm_limit: Optional[dict] - key-specific per-request-tag rpm limit, keyed by request tag. Example - {"cell-1": 1000, "cell-2": 500}. Each tag gets an independent counter; requests whose tag is absent fall back to the key-level rpm limit. - tpm_limit_type: Optional[str] - Type of tpm limit. Options: "best_effort_throughput" (no error if we're overallocating tpm), "guaranteed_throughput" (raise an error if we're overallocating tpm), "dynamic" (dynamically exceed limit when no 429 errors). Defaults to "best_effort_throughput". - rpm_limit_type: Optional[str] - Type of rpm limit. Options: "best_effort_throughput" (no error if we're overallocating rpm), "guaranteed_throughput" (raise an error if we're overallocating rpm), "dynamic" (dynamically exceed limit when no 429 errors). Defaults to "best_effort_throughput". - allowed_cache_controls: Optional[list] - List of allowed cache control values. Example - ["no-cache", "no-store"]. See all values - https://docs.litellm.ai/docs/proxy/caching#turn-on--off-caching-per-request - blocked: Optional[bool] - Whether the key is blocked. - rpm_limit: Optional[int] - Specify rpm limit for a given key (Requests per minute) - tpm_limit: Optional[int] - Specify tpm limit for a given key (Tokens per minute) - soft_budget: Optional[float] - Specify soft budget for a given key. Will trigger a slack alert when this soft budget is reached. - tags: Optional[List[str]] - Tags for [tracking spend](https://litellm.vercel.app/docs/proxy/enterprise#tracking-spend-for-custom-tags) and/or doing [tag-based routing](https://litellm.vercel.app/docs/proxy/tag_routing). - prompts: Optional[List[str]] - List of prompts that the key is allowed to use. - enforced_params: Optional[List[str]] - List of enforced params for the key (Enterprise only). [Docs](https://docs.litellm.ai/docs/proxy/enterprise#enforce-required-params-for-llm-requests) - prompts: Optional[List[str]] - List of prompts that the key is allowed to use. - allowed_routes: Optional[list] - List of allowed routes for the key. Store the actual route or store a wildcard pattern for a set of routes. Example - ["/chat/completions", "/embeddings", "/keys/*"] - allowed_passthrough_routes: Optional[list] - List of allowed pass through endpoints for the key. Store the actual endpoint or store a wildcard pattern for a set of endpoints. Example - ["/my-custom-endpoint"]. Use this instead of allowed_routes, if you just want to specify which pass through endpoints the key can access, without specifying the routes. If allowed_routes is specified, allowed_pass_through_endpoints is ignored. - object_permission: Optional[LiteLLM_ObjectPermissionBase] - key-specific object permission. Example - {"vector_stores": ["vector_store_1", "vector_store_2"], "agents": ["agent_1", "agent_2"], "agent_access_groups": ["dev_group"]}. IF null or {} then no object permission. - key_type: Optional[str] - Type of key that determines default allowed routes. Options: "llm_api" (can call LLM API routes), "management" (can call management routes), "read_only" (can only call info/read routes), "default" (uses default allowed routes). Defaults to "default". - prompts: Optional[List[str]] - List of allowed prompts for the key. If specified, the key will only be able to use these specific prompts. - auto_rotate: Optional[bool] - Whether this key should be automatically rotated (regenerated) - rotation_interval: Optional[str] - How often to auto-rotate this key (e.g., '30s', '30m', '30h', '30d'). Required if auto_rotate=True. - allowed_vector_store_indexes: Optional[List[dict]] - List of allowed vector store indexes for the key. Example - [{"index_name": "my-index", "index_permissions": ["write", "read"]}]. If specified, the key will only be able to use these specific vector store indexes. Create index, using `/v1/indexes` endpoint. - router_settings: Optional[UpdateRouterConfig] - key-specific router settings. Example - {"model_group_retry_policy": {"gpt-4": {"RateLimitErrorRetries": 5}}}. IF null or {} then no router settings. - access_group_ids: Optional[List[str]] - List of access group IDs to associate with the key. Access groups define which models a key can access. Example - ["access_group_1", "access_group_2"]. - budget_limits: Optional[list] - List of concurrent budget windows for the key. Each window specifies a budget_limit, time_period, and optional budget_duration. Example - [{"budget_limit": 10.0, "time_period": "1d"}, {"budget_limit": 50.0, "time_period": "7d"}]. Examples: 1. Allow users to turn on/off pii masking ```bash curl --location 'http://0.0.0.0:4000/key/generate' --header 'Authorization: Bearer sk-1234' --header 'Content-Type: application/json' --data '{ "permissions": {"allow_pii_controls": true} }' ``` Returns: - key: (str) The generated api key - expires: (datetime) Datetime object for when key expires. - user_id: (str) Unique user id - used for tracking spend across multiple keys for same user id. */
+/** Generate Key Fn Generate an API key based on the provided data. Docs: https://docs.litellm.ai/docs/proxy/virtual_keys Parameters: - duration: Optional[str] - Specify the length of time the token is valid for. You can set duration as seconds ("30s"), minutes ("30m"), hours ("30h"), days ("30d"). - key_alias: Optional[str] - User defined key alias - key: Optional[str] - User defined key value. Must start with 'sk-' and be at least 16 characters long. If not set, a 16-digit unique sk-key is created for you. - team_id: Optional[str] - The team id of the key - user_id: Optional[str] - The user id of the key - agent_id: Optional[str] - The agent id associated with the key. - organization_id: Optional[str] - The organization id of the key. If not set, and team_id is set, the organization id will be the same as the team id. If conflict, an error will be raised. - project_id: Optional[str] - The project id of the key. When set, models and max_budget are validated against the project's limits. - budget_id: Optional[str] - The budget id associated with the key. Created by calling `/budget/new`. - end_user_budget_id: Optional[str] - Proxy admin only. Budget id applied to end users first seen through this key that carry no budget of their own. Takes precedence over `litellm_settings.max_end_user_budget_id`. - models: Optional[list] - Model_name's a user is allowed to call. (if empty, key is allowed to call all models) - aliases: Optional[dict] - Any alias mappings, on top of anything in the config.yaml model list. - https://docs.litellm.ai/docs/proxy/virtual_keys#managing-auth---upgradedowngrade-models - config: Optional[dict] - any key-specific configs, overrides config in config.yaml - spend: Optional[int] - Amount spent by key. Default is 0. Will be updated by proxy whenever key is used. https://docs.litellm.ai/docs/proxy/virtual_keys#managing-auth---tracking-spend - send_invite_email: Optional[bool] - Whether to send an invite email to the user_id, with the generate key - max_budget: Optional[float] - Specify max budget for a given key. - budget_duration: Optional[str] - Budget is reset at the end of specified duration. If not set, budget is never reset. You can set duration as seconds ("30s"), minutes ("30m"), hours ("30h"), days ("30d"). - max_parallel_requests: Optional[int] - Rate limit a user based on the number of parallel requests. Raises 429 error, if user's parallel requests > x. - metadata: Optional[dict] - Metadata for key, store information for key. Example metadata = {"team": "core-infra", "app": "app2", "email": "ishaan@berri.ai" } - guardrails: Optional[List[str]] - List of active guardrails for the key - policies: Optional[List[str]] - List of policy names to apply to the key. Policies define guardrails, conditions, and inheritance rules. - disable_global_guardrails: Optional[bool] - Whether to disable global guardrails for the key. - throttle_on_budget_exceeded: Optional[bool] - When the key exceeds its max_budget, throttle its tpm/rpm to the global budget_exceeded_throttle_percentage instead of blocking the key entirely. - enable_prompt_caching: Optional[bool] - Auto-inject prompt caching breakpoints (Anthropic cache_control markers) on requests made with this key. Supported Claude models on Anthropic, Bedrock, Vertex AI, and Azure AI only. - permissions: Optional[dict] - key-specific permissions. Currently just used for turning off pii masking (if connected). Example - {"pii": false} - model_max_budget: Optional[Dict[str, BudgetConfig]] - Model-specific budgets {"gpt-4": {"budget_limit": 0.0005, "time_period": "30d"}}}. IF null or {} then no model specific budget. - budget_fallbacks: Optional[Dict[str, List[str]]] - Per-model fallback chain tried in order when that model's own `model_max_budget` is exceeded, e.g. {"gpt-4o": ["gpt-4o-mini"]}. - model_rpm_limit: Optional[dict] - key-specific model rpm limit. Example - {"text-davinci-002": 1000, "gpt-3.5-turbo": 1000}. IF null or {} then no model specific rpm limit. - model_tpm_limit: Optional[dict] - key-specific model tpm limit. Example - {"text-davinci-002": 1000, "gpt-3.5-turbo": 1000}. IF null or {} then no model specific tpm limit. - default_estimated_output_tokens: Optional[int] - Proxy admin only. Expected output tokens reserved for TPM limiting when a request omits max_tokens. Positive integer. Falls back to the team setting, then to the built-in estimate. - default_estimated_output_tokens_per_model: Optional[dict] - Proxy admin only. Per-model override of the above. Example - {"gpt-4": 4096, "gpt-3.5-turbo": 1024}. Takes precedence over the key-wide value. - mcp_rpm_limit: Optional[dict] - key-specific per-MCP-server rpm limit, keyed by MCP server name (alias if set, else the configured name). Example - {"github": 100, "slack": 200}. IF null or {} then no MCP-specific rpm limit. - tag_rpm_limit: Optional[dict] - key-specific per-request-tag rpm limit, keyed by request tag. Example - {"cell-1": 1000, "cell-2": 500}. Each tag gets an independent counter; requests whose tag is absent fall back to the key-level rpm limit. - tpm_limit_type: Optional[str] - Type of tpm limit. Options: "best_effort_throughput" (no error if we're overallocating tpm), "guaranteed_throughput" (raise an error if we're overallocating tpm), "dynamic" (dynamically exceed limit when no 429 errors). Defaults to "best_effort_throughput". - rpm_limit_type: Optional[str] - Type of rpm limit. Options: "best_effort_throughput" (no error if we're overallocating rpm), "guaranteed_throughput" (raise an error if we're overallocating rpm), "dynamic" (dynamically exceed limit when no 429 errors). Defaults to "best_effort_throughput". - allowed_cache_controls: Optional[list] - List of allowed cache control values. Example - ["no-cache", "no-store"]. See all values - https://docs.litellm.ai/docs/proxy/caching#turn-on--off-caching-per-request - blocked: Optional[bool] - Whether the key is blocked. - rpm_limit: Optional[int] - Specify rpm limit for a given key (Requests per minute) - tpm_limit: Optional[int] - Specify tpm limit for a given key (Tokens per minute) - tpd_limit: Optional[int] - Specify tpd limit for a given key (Tokens per day). Charged by batch submissions instead of tpm_limit/rpm_limit. - soft_budget: Optional[float] - Specify soft budget for a given key. Will trigger a slack alert when this soft budget is reached. - tags: Optional[List[str]] - Tags for [tracking spend](https://litellm.vercel.app/docs/proxy/enterprise#tracking-spend-for-custom-tags) and/or doing [tag-based routing](https://litellm.vercel.app/docs/proxy/tag_routing). - prompts: Optional[List[str]] - List of prompts that the key is allowed to use. - enforced_params: Optional[List[str]] - List of enforced params for the key (Enterprise only). [Docs](https://docs.litellm.ai/docs/proxy/enterprise#enforce-required-params-for-llm-requests) - prompts: Optional[List[str]] - List of prompts that the key is allowed to use. - allowed_routes: Optional[list] - List of allowed routes for the key. Store the actual route or store a wildcard pattern for a set of routes. Example - ["/chat/completions", "/embeddings", "/keys/*"] - allowed_passthrough_routes: Optional[list] - List of allowed pass through endpoints for the key. Store the actual endpoint or store a wildcard pattern for a set of endpoints. Example - ["/my-custom-endpoint"]. Use this instead of allowed_routes, if you just want to specify which pass through endpoints the key can access, without specifying the routes. If allowed_routes is specified, allowed_pass_through_endpoints is ignored. - object_permission: Optional[LiteLLM_ObjectPermissionBase] - key-specific object permission. Example - {"vector_stores": ["vector_store_1", "vector_store_2"], "agents": ["agent_1", "agent_2"], "agent_access_groups": ["dev_group"]}. IF null or {} then no object permission. - key_type: Optional[str] - Type of key that determines default allowed routes. Options: "llm_api" (can call LLM API routes), "management" (can call management routes), "read_only" (can only call info/read routes), "default" (uses default allowed routes). Defaults to "default". - prompts: Optional[List[str]] - List of allowed prompts for the key. If specified, the key will only be able to use these specific prompts. - auto_rotate: Optional[bool] - Whether this key should be automatically rotated (regenerated) - rotation_interval: Optional[str] - How often to auto-rotate this key (e.g., '30s', '30m', '30h', '30d'). Required if auto_rotate=True. - allowed_vector_store_indexes: Optional[List[dict]] - List of allowed vector store indexes for the key. Example - [{"index_name": "my-index", "index_permissions": ["write", "read"]}]. If specified, the key will only be able to use these specific vector store indexes. Create index, using `/v1/indexes` endpoint. - router_settings: Optional[UpdateRouterConfig] - key-specific router settings. Example - {"model_group_retry_policy": {"gpt-4": {"RateLimitErrorRetries": 5}}}. IF null or {} then no router settings. - access_group_ids: Optional[List[str]] - List of access group IDs to associate with the key. Access groups define which models a key can access. Example - ["access_group_1", "access_group_2"]. - budget_limits: Optional[list] - List of concurrent budget windows for the key. Each window specifies a budget_limit, time_period, and optional budget_duration. Example - [{"budget_limit": 10.0, "time_period": "1d"}, {"budget_limit": 50.0, "time_period": "7d"}]. Examples: 1. Allow users to turn on/off pii masking ```bash curl --location 'http://0.0.0.0:4000/key/generate' --header 'Authorization: Bearer sk-1234' --header 'Content-Type: application/json' --data '{ "permissions": {"allow_pii_controls": true} }' ``` Returns: - key: (str) The generated api key - expires: (datetime) Datetime object for when key expires. - user_id: (str) Unique user id - used for tracking spend across multiple keys for same user id. */
 export const generateKeyFnKeyGeneratePost: API.OperationMethod<
   GenerateKeyFnKeyGeneratePostRequest,
   GenerateKeyResponse,
@@ -3964,7 +4094,7 @@ export const generateKeyFnKeyGeneratePost: API.OperationMethod<
 export type GenerateServiceAccountKeyFnKeyServiceAccountGeneratePostError =
   | UnprocessableEntity
   | LitellmOpError;
-/** Generate Service Account Key Fn Generate a Service Account API key based on the provided data. This key does not belong to any user. It belongs to the team. Why use a service account key? - Prevent key from being deleted when user is deleted. - Apply team limits, not team member limits to key. Docs: https://docs.litellm.ai/docs/proxy/virtual_keys Parameters: - duration: Optional[str] - Specify the length of time the token is valid for. You can set duration as seconds ("30s"), minutes ("30m"), hours ("30h"), days ("30d"). - key_alias: Optional[str] - User defined key alias - key: Optional[str] - User defined key value. Must start with 'sk-' and be at least 16 characters long. If not set, a 16-digit unique sk-key is created for you. - team_id: Optional[str] - The team id of the key - user_id: Optional[str] - [NON-FUNCTIONAL] THIS WILL BE IGNORED. The user id of the key - budget_id: Optional[str] - The budget id associated with the key. Created by calling `/budget/new`. - models: Optional[list] - Model_name's a user is allowed to call. (if empty, key is allowed to call all models) - aliases: Optional[dict] - Any alias mappings, on top of anything in the config.yaml model list. - https://docs.litellm.ai/docs/proxy/virtual_keys#managing-auth---upgradedowngrade-models - config: Optional[dict] - any key-specific configs, overrides config in config.yaml - spend: Optional[int] - Amount spent by key. Default is 0. Will be updated by proxy whenever key is used. https://docs.litellm.ai/docs/proxy/virtual_keys#managing-auth---tracking-spend - send_invite_email: Optional[bool] - Whether to send an invite email to the user_id, with the generate key - max_budget: Optional[float] - Specify max budget for a given key. - budget_duration: Optional[str] - Budget is reset at the end of specified duration. If not set, budget is never reset. You can set duration as seconds ("30s"), minutes ("30m"), hours ("30h"), days ("30d"). - max_parallel_requests: Optional[int] - Rate limit a user based on the number of parallel requests. Raises 429 error, if user's parallel requests > x. - metadata: Optional[dict] - Metadata for key, store information for key. Example metadata = {"team": "core-infra", "app": "app2", "email": "ishaan@berri.ai" } - guardrails: Optional[List[str]] - List of active guardrails for the key - permissions: Optional[dict] - key-specific permissions. Currently just used for turning off pii masking (if connected). Example - {"pii": false} - model_max_budget: Optional[Dict[str, BudgetConfig]] - Model-specific budgets {"gpt-4": {"budget_limit": 0.0005, "time_period": "30d"}}}. IF null or {} then no model specific budget. - budget_fallbacks: Optional[Dict[str, List[str]]] - Per-model fallback chain tried in order when that model's own `model_max_budget` is exceeded, e.g. {"gpt-4o": ["gpt-4o-mini"]}. - model_rpm_limit: Optional[dict] - key-specific model rpm limit. Example - {"text-davinci-002": 1000, "gpt-3.5-turbo": 1000}. IF null or {} then no model specific rpm limit. - model_tpm_limit: Optional[dict] - key-specific model tpm limit. Example - {"text-davinci-002": 1000, "gpt-3.5-turbo": 1000}. IF null or {} then no model specific tpm limit. - default_estimated_output_tokens: Optional[int] - Proxy admin only. Expected output tokens reserved for TPM limiting when a request omits max_tokens. Positive integer. Falls back to the team setting, then to the built-in estimate. - default_estimated_output_tokens_per_model: Optional[dict] - Proxy admin only. Per-model override of the above. Example - {"gpt-4": 4096, "gpt-3.5-turbo": 1024}. Takes precedence over the key-wide value. - mcp_rpm_limit: Optional[dict] - key-specific per-MCP-server rpm limit, keyed by MCP server name (alias if set, else the configured name). Example - {"github": 100, "slack": 200}. IF null or {} then no MCP-specific rpm limit. - tpm_limit_type: Optional[str] - TPM rate limit type - "best_effort_throughput", "guaranteed_throughput", or "dynamic" - rpm_limit_type: Optional[str] - RPM rate limit type - "best_effort_throughput", "guaranteed_throughput", or "dynamic" - allowed_cache_controls: Optional[list] - List of allowed cache control values. Example - ["no-cache", "no-store"]. See all values - https://docs.litellm.ai/docs/proxy/caching#turn-on--off-caching-per-request - blocked: Optional[bool] - Whether the key is blocked. - rpm_limit: Optional[int] - Specify rpm limit for a given key (Requests per minute) - tpm_limit: Optional[int] - Specify tpm limit for a given key (Tokens per minute) - soft_budget: Optional[float] - Specify soft budget for a given key. Will trigger a slack alert when this soft budget is reached. - tags: Optional[List[str]] - Tags for [tracking spend](https://litellm.vercel.app/docs/proxy/enterprise#tracking-spend-for-custom-tags) and/or doing [tag-based routing](https://litellm.vercel.app/docs/proxy/tag_routing). - enforced_params: Optional[List[str]] - List of enforced params for the key (Enterprise only). [Docs](https://docs.litellm.ai/docs/proxy/enterprise#enforce-required-params-for-llm-requests) - allowed_routes: Optional[list] - List of allowed routes for the key. Store the actual route or store a wildcard pattern for a set of routes. Example - ["/chat/completions", "/embeddings", "/keys/*"] - object_permission: Optional[LiteLLM_ObjectPermissionBase] - key-specific object permission. Example - {"vector_stores": ["vector_store_1", "vector_store_2"], "agents": ["agent_1", "agent_2"], "agent_access_groups": ["dev_group"]}. IF null or {} then no object permission. Examples: - allowed_vector_store_indexes: Optional[List[dict]] - List of allowed vector store indexes for the key. Example - [{"index_name": "my-index", "index_permissions": ["write", "read"]}]. If specified, the key will only be able to use these specific vector store indexes. Create index, using `/v1/indexes` endpoint. 1. Allow users to turn on/off pii masking ```bash curl --location 'http://0.0.0.0:4000/key/generate' --header 'Authorization: Bearer sk-1234' --header 'Content-Type: application/json' --data '{ "permissions": {"allow_pii_controls": true} }' ``` Returns: - key: (str) The generated api key - expires: (datetime) Datetime object for when key expires. - user_id: (str) Unique user id - used for tracking spend across multiple keys for same user id. */
+/** Generate Service Account Key Fn Generate a Service Account API key based on the provided data. This key does not belong to any user. It belongs to the team. Why use a service account key? - Prevent key from being deleted when user is deleted. - Apply team limits, not team member limits to key. Docs: https://docs.litellm.ai/docs/proxy/virtual_keys Parameters: - duration: Optional[str] - Specify the length of time the token is valid for. You can set duration as seconds ("30s"), minutes ("30m"), hours ("30h"), days ("30d"). - key_alias: Optional[str] - User defined key alias - key: Optional[str] - User defined key value. Must start with 'sk-' and be at least 16 characters long. If not set, a 16-digit unique sk-key is created for you. - team_id: Optional[str] - The team id of the key - user_id: Optional[str] - [NON-FUNCTIONAL] THIS WILL BE IGNORED. The user id of the key - budget_id: Optional[str] - The budget id associated with the key. Created by calling `/budget/new`. - end_user_budget_id: Optional[str] - Proxy admin only. Budget id applied to end users first seen through this key that carry no budget of their own. Omit to keep the current value, pass an empty string to clear it. - models: Optional[list] - Model_name's a user is allowed to call. (if empty, key is allowed to call all models) - aliases: Optional[dict] - Any alias mappings, on top of anything in the config.yaml model list. - https://docs.litellm.ai/docs/proxy/virtual_keys#managing-auth---upgradedowngrade-models - config: Optional[dict] - any key-specific configs, overrides config in config.yaml - spend: Optional[int] - Amount spent by key. Default is 0. Will be updated by proxy whenever key is used. https://docs.litellm.ai/docs/proxy/virtual_keys#managing-auth---tracking-spend - send_invite_email: Optional[bool] - Whether to send an invite email to the user_id, with the generate key - max_budget: Optional[float] - Specify max budget for a given key. - budget_duration: Optional[str] - Budget is reset at the end of specified duration. If not set, budget is never reset. You can set duration as seconds ("30s"), minutes ("30m"), hours ("30h"), days ("30d"). - max_parallel_requests: Optional[int] - Rate limit a user based on the number of parallel requests. Raises 429 error, if user's parallel requests > x. - metadata: Optional[dict] - Metadata for key, store information for key. Example metadata = {"team": "core-infra", "app": "app2", "email": "ishaan@berri.ai" } - guardrails: Optional[List[str]] - List of active guardrails for the key - permissions: Optional[dict] - key-specific permissions. Currently just used for turning off pii masking (if connected). Example - {"pii": false} - model_max_budget: Optional[Dict[str, BudgetConfig]] - Model-specific budgets {"gpt-4": {"budget_limit": 0.0005, "time_period": "30d"}}}. IF null or {} then no model specific budget. - budget_fallbacks: Optional[Dict[str, List[str]]] - Per-model fallback chain tried in order when that model's own `model_max_budget` is exceeded, e.g. {"gpt-4o": ["gpt-4o-mini"]}. - model_rpm_limit: Optional[dict] - key-specific model rpm limit. Example - {"text-davinci-002": 1000, "gpt-3.5-turbo": 1000}. IF null or {} then no model specific rpm limit. - model_tpm_limit: Optional[dict] - key-specific model tpm limit. Example - {"text-davinci-002": 1000, "gpt-3.5-turbo": 1000}. IF null or {} then no model specific tpm limit. - default_estimated_output_tokens: Optional[int] - Proxy admin only. Expected output tokens reserved for TPM limiting when a request omits max_tokens. Positive integer. Falls back to the team setting, then to the built-in estimate. - default_estimated_output_tokens_per_model: Optional[dict] - Proxy admin only. Per-model override of the above. Example - {"gpt-4": 4096, "gpt-3.5-turbo": 1024}. Takes precedence over the key-wide value. - mcp_rpm_limit: Optional[dict] - key-specific per-MCP-server rpm limit, keyed by MCP server name (alias if set, else the configured name). Example - {"github": 100, "slack": 200}. IF null or {} then no MCP-specific rpm limit. - tpm_limit_type: Optional[str] - TPM rate limit type - "best_effort_throughput", "guaranteed_throughput", or "dynamic" - rpm_limit_type: Optional[str] - RPM rate limit type - "best_effort_throughput", "guaranteed_throughput", or "dynamic" - allowed_cache_controls: Optional[list] - List of allowed cache control values. Example - ["no-cache", "no-store"]. See all values - https://docs.litellm.ai/docs/proxy/caching#turn-on--off-caching-per-request - blocked: Optional[bool] - Whether the key is blocked. - rpm_limit: Optional[int] - Specify rpm limit for a given key (Requests per minute) - tpm_limit: Optional[int] - Specify tpm limit for a given key (Tokens per minute) - tpd_limit: Optional[int] - Specify tpd limit for a given key (Tokens per day). Charged by batch submissions instead of tpm_limit/rpm_limit. - soft_budget: Optional[float] - Specify soft budget for a given key. Will trigger a slack alert when this soft budget is reached. - tags: Optional[List[str]] - Tags for [tracking spend](https://litellm.vercel.app/docs/proxy/enterprise#tracking-spend-for-custom-tags) and/or doing [tag-based routing](https://litellm.vercel.app/docs/proxy/tag_routing). - enforced_params: Optional[List[str]] - List of enforced params for the key (Enterprise only). [Docs](https://docs.litellm.ai/docs/proxy/enterprise#enforce-required-params-for-llm-requests) - allowed_routes: Optional[list] - List of allowed routes for the key. Store the actual route or store a wildcard pattern for a set of routes. Example - ["/chat/completions", "/embeddings", "/keys/*"] - object_permission: Optional[LiteLLM_ObjectPermissionBase] - key-specific object permission. Example - {"vector_stores": ["vector_store_1", "vector_store_2"], "agents": ["agent_1", "agent_2"], "agent_access_groups": ["dev_group"]}. IF null or {} then no object permission. Examples: - allowed_vector_store_indexes: Optional[List[dict]] - List of allowed vector store indexes for the key. Example - [{"index_name": "my-index", "index_permissions": ["write", "read"]}]. If specified, the key will only be able to use these specific vector store indexes. Create index, using `/v1/indexes` endpoint. 1. Allow users to turn on/off pii masking ```bash curl --location 'http://0.0.0.0:4000/key/generate' --header 'Authorization: Bearer sk-1234' --header 'Content-Type: application/json' --data '{ "permissions": {"allow_pii_controls": true} }' ``` Returns: - key: (str) The generated api key - expires: (datetime) Datetime object for when key expires. - user_id: (str) Unique user id - used for tracking spend across multiple keys for same user id. */
 export const generateServiceAccountKeyFnKeyServiceAccountGeneratePost: API.OperationMethod<
   GenerateServiceAccountKeyFnKeyServiceAccountGeneratePostRequest,
   GenerateServiceAccountKeyFnKeyServiceAccountGeneratePostResponse,
@@ -3979,7 +4109,7 @@ export const generateServiceAccountKeyFnKeyServiceAccountGeneratePost: API.Opera
 }));
 
 export type GetInfoKeyFnKeyInfoError = UnprocessableEntity | LitellmOpError;
-/** Info Key Fn Retrieve information about a key. Parameters: - key: str | None (query parameter) - The key to look up. Accepts the plaintext key or its hash. Defaults to the key in the Authorization header. Returns: - key: str - The key that was looked up, echoed back as it was passed in - info: dict - The key's row, minus the hashed token - key_alias: str | None - User-friendly key alias - spend: float - Amount spent by the key. When budget_duration is set this covers only the current budget window, not the key's lifetime - max_budget: float | None - Max budget for the key, enforced against spend - budget_duration: str | None - Budget reset period ("30d", "1h", etc.) - budget_reset_at: datetime | None - When the current budget window ends and spend is next reset to 0, not when it was last reset. Reset times snap to standard boundaries in the configured timezone (30d and 1mo land on the 1st of the month, 7d on Monday, 1h on the hour), so subtracting budget_duration from it does not give the window's start - model_max_budget: dict - Per-model budgets, e.g. {"gpt-4": {"budget_limit": 0.0005, "time_period": "30d"}} - model_max_budget_usage: dict | None - Current-window spend per model, present only when the key has per-model budgets - models: list - Model_name's the key is allowed to call - tpm_limit / rpm_limit: int | None - Tokens and requests per minute limits - metadata: dict - Metadata for the key, e.g. {"team": "core-infra"} - blocked: bool | None - Whether the key is blocked - expires: datetime | None - When the key stops authenticating requests - last_active: datetime | None - When the key was last used - object_permission: dict | None - Resolved vector store / MCP permissions when the key has an object_permission_id Example Curl: ``` curl -X GET "http://0.0.0.0:4000/key/info?key=sk-test-example-key-123" -H "Authorization: Bearer sk-1234" ``` Example Curl - if no key is passed, it will use the Key Passed in Authorization Header ``` curl -X GET "http://0.0.0.0:4000/key/info" -H "Authorization: Bearer sk-test-example-key-123" ``` */
+/** Info Key Fn Retrieve information about a key. Parameters: - key: str | None (query parameter) - The key to look up. Accepts the plaintext key or its hash; prefer the hash, since a query parameter is recorded verbatim by any HTTP access log in front of the proxy. Defaults to the key in the Authorization header. Returns: - key: str - The key that was looked up, echoed back as it was passed in - info: dict - The key's row, minus the hashed token. Deleted keys are served from the LiteLLM_DeletedVerificationToken archive and carry deleted_at / deleted_by - status: "active" | "expired" | "revoked" | "deleted" - Derived from blocked, expires and whether the row came from the archive - key_alias: str | None - User-friendly key alias - spend: float - Amount spent by the key. When budget_duration is set this covers only the current budget window, not the key's lifetime - max_budget: float | None - Max budget for the key, enforced against spend - budget_duration: str | None - Budget reset period ("30d", "1h", etc.) - budget_reset_at: datetime | None - When the current budget window ends and spend is next reset to 0, not when it was last reset. Reset times snap to standard boundaries in the configured timezone (30d and 1mo land on the 1st of the month, 7d on Monday, 1h on the hour), so subtracting budget_duration from it does not give the window's start - model_max_budget: dict - Per-model budgets, e.g. {"gpt-4": {"budget_limit": 0.0005, "time_period": "30d"}} - model_max_budget_usage: dict | None - Current-window spend per model, present only when the key has per-model budgets - budget_limits: list | None - Concurrent budget windows, exactly as stored - budget_limits_usage: dict | None - Current-window spend per budget window, e.g. {"1h": {"current_spend": 0.0009}}, present only when the key has budget windows (read from the same cross-pod spend counter the budget enforcement uses) - models: list - Model_name's the key is allowed to call - tpm_limit / rpm_limit: int | None - Tokens and requests per minute limits - metadata: dict - Metadata for the key, e.g. {"team": "core-infra"} - blocked: bool | None - Whether the key is blocked - expires: datetime | None - When the key stops authenticating requests - last_active: datetime | None - When the key was last used - object_permission: dict | None - Resolved vector store / MCP permissions when the key has an object_permission_id Example Curl: ``` curl -X GET "http://0.0.0.0:4000/key/info?key=d5345c0ecc68ae6295c69f91926b2bd379e25481a40c34b5884d157a9f65d8fa" -H "Authorization: Bearer sk-1234" ``` Example Curl - if no key is passed, it will use the Key Passed in Authorization Header ``` curl -X GET "http://0.0.0.0:4000/key/info" -H "Authorization: Bearer sk-test-example-key-123" ``` */
 export const getInfoKeyFnKeyInfo: API.OperationMethod<
   GetInfoKeyFnKeyInfoRequest,
   GetInfoKeyFnKeyInfoResponse,
@@ -4012,7 +4142,7 @@ export type ListKeysKeyListGetError =
   | BadRequest
   | UnprocessableEntity
   | LitellmOpError;
-/** List Keys List all keys for a given user / team / organization. Parameters: expand: Optional[List[str]] - Expand related objects (e.g. 'user' to include user information) status: Optional[str] - Filter by status. Currently supports "deleted" to query deleted keys. Returns: { "keys": List[str] or List[UserAPIKeyAuth], "total_count": int, "current_page": int, "total_pages": int, } When expand includes "user", each key object will include a "user" field with the associated user object. Note: When expand=user is specified, full key objects are returned regardless of the return_full_object parameter. */
+/** List Keys List all keys for a given user / team / organization. Parameters: expand: Optional[List[str]] - Expand related objects (e.g. 'user' to include user information) status: Optional[str] - Filter by status: "active", "expired", "revoked" (blocked) or "deleted". "deleted" reads the LiteLLM_DeletedVerificationToken archive; the other values partition the live key table, so every live key matches exactly one of them. Returns: { "keys": List[str] or List[UserAPIKeyAuth], "total_count": int, "current_page": int, "total_pages": int, } When expand includes "user", each key object will include a "user" field with the associated user object. Note: When expand=user is specified, full key objects are returned regardless of the return_full_object parameter. */
 export const listKeysKeyListGet: API.OperationMethod<
   ListKeysKeyListGetRequest,
   KeyListResponseObject,
@@ -4147,7 +4277,7 @@ export type UpdateKeyFnKeyUpdatePostError =
   | NotFound
   | UnprocessableEntity
   | LitellmOpError;
-/** Update Key Fn Update an existing API key's parameters. Parameters: - key: Optional[str] - The key to update. Either key or key_alias must be provided. - key_alias: Optional[str] - User-friendly key alias. If key is omitted, also identifies the key to update (must match exactly one key, same as /key/delete's key_aliases) - user_id: Optional[str] - User ID associated with key - team_id: Optional[str] - Team ID associated with key - agent_id: Optional[str] - The agent id associated with the key. - organization_id: Optional[str] - The organization id of the key. - budget_id: Optional[str] - The budget id associated with the key. Created by calling `/budget/new`. - models: Optional[list] - Model_name's a user is allowed to call - tags: Optional[List[str]] - Tags for organizing keys (Enterprise only) - prompts: Optional[List[str]] - List of prompts that the key is allowed to use. - enforced_params: Optional[List[str]] - List of enforced params for the key (Enterprise only). [Docs](https://docs.litellm.ai/docs/proxy/enterprise#enforce-required-params-for-llm-requests) - spend: Optional[float] - Amount spent by key - max_budget: Optional[float] - Max budget for key - model_max_budget: Optional[Dict[str, BudgetConfig]] - Model-specific budgets {"gpt-4": {"budget_limit": 0.0005, "time_period": "30d"}} - budget_fallbacks: Optional[Dict[str, List[str]]] - Per-model fallback chain tried in order when that model's own `model_max_budget` is exceeded, e.g. {"gpt-4o": ["gpt-4o-mini"]}. - budget_duration: Optional[str] - Budget reset period ("30d", "1h", etc.) - soft_budget: Optional[float] - [TODO] Soft budget limit (warning vs. hard stop). Will trigger a slack alert when this soft budget is reached. - max_parallel_requests: Optional[int] - Rate limit for parallel requests - metadata: Optional[dict] - Metadata for key. Example {"team": "core-infra", "app": "app2"} - tpm_limit: Optional[int] - Tokens per minute limit - rpm_limit: Optional[int] - Requests per minute limit - model_rpm_limit: Optional[dict] - Model-specific RPM limits {"gpt-4": 100, "claude-v1": 200} - mcp_rpm_limit: Optional[dict] - Per-MCP-server RPM limits, keyed by MCP server name {"github": 100, "slack": 200} - tag_rpm_limit: Optional[dict] - Per-request-tag RPM limits, keyed by request tag {"cell-1": 1000, "cell-2": 500}. Each tag gets an independent counter; absent tags fall back to the key-level rpm limit. - model_tpm_limit: Optional[dict] - Model-specific TPM limits {"gpt-4": 100000, "claude-v1": 200000} - default_estimated_output_tokens: Optional[int] - Proxy admin only. Expected output tokens reserved for TPM limiting when a request omits max_tokens. Positive integer. - default_estimated_output_tokens_per_model: Optional[dict] - Proxy admin only. Per-model override of the above {"gpt-4": 4096, "gpt-3.5-turbo": 1024} - tpm_limit_type: Optional[str] - TPM rate limit type - "best_effort_throughput", "guaranteed_throughput", or "dynamic" - rpm_limit_type: Optional[str] - RPM rate limit type - "best_effort_throughput", "guaranteed_throughput", or "dynamic" - allowed_cache_controls: Optional[list] - List of allowed cache control values - duration: Optional[str] - Key validity duration ("30d", "1h", etc.), null to never expire, or "-1" to never expire (deprecated, use null) - permissions: Optional[dict] - Key-specific permissions - send_invite_email: Optional[bool] - Send invite email to user_id - guardrails: Optional[List[str]] - List of active guardrails for the key - policies: Optional[List[str]] - List of policy names to apply to the key. Policies define guardrails, conditions, and inheritance rules. - disable_global_guardrails: Optional[bool] - Whether to disable global guardrails for the key. - throttle_on_budget_exceeded: Optional[bool] - When the key exceeds its max_budget, throttle its tpm/rpm to the global budget_exceeded_throttle_percentage instead of blocking the key entirely. - enable_prompt_caching: Optional[bool] - Auto-inject prompt caching breakpoints (Anthropic cache_control markers) on requests made with this key. Anthropic and Bedrock Claude models only. - prompts: Optional[List[str]] - List of prompts that the key is allowed to use. - blocked: Optional[bool] - Whether the key is blocked - aliases: Optional[dict] - Model aliases for the key - [Docs](https://litellm.vercel.app/docs/proxy/virtual_keys#model-aliases) - config: Optional[dict] - [DEPRECATED PARAM] Key-specific config. - temp_budget_increase: Optional[float] - Temporary budget increase for the key (Enterprise only). - temp_budget_expiry: Optional[str] - Expiry time for the temporary budget increase (Enterprise only). - allowed_routes: Optional[list] - List of allowed routes for the key. Store the actual route or store a wildcard pattern for a set of routes. Example - ["/chat/completions", "/embeddings", "/keys/*"] - allowed_passthrough_routes: Optional[list] - List of allowed pass through routes for the key. Store the actual route or store a wildcard pattern for a set of routes. Example - ["/my-custom-endpoint"]. Use this instead of allowed_routes, if you just want to specify which pass through routes the key can access, without specifying the routes. If allowed_routes is specified, allowed_passthrough_routes is ignored. - prompts: Optional[List[str]] - List of allowed prompts for the key. If specified, the key will only be able to use these specific prompts. - object_permission: Optional[LiteLLM_ObjectPermissionBase] - key-specific object permission. Example - {"vector_stores": ["vector_store_1", "vector_store_2"], "agents": ["agent_1", "agent_2"], "agent_access_groups": ["dev_group"]}. IF null or {} then no object permission. - auto_rotate: Optional[bool] - Whether this key should be automatically rotated - rotation_interval: Optional[str] - How often to rotate this key (e.g., '30d', '90d'). Required if auto_rotate=True - allowed_vector_store_indexes: Optional[List[dict]] - List of allowed vector store indexes for the key. Example - [{"index_name": "my-index", "index_permissions": ["write", "read"]}]. If specified, the key will only be able to use these specific vector store indexes. Create index, using `/v1/indexes` endpoint. - router_settings: Optional[UpdateRouterConfig] - key-specific router settings. Example - {"model_group_retry_policy": {"gpt-4": {"RateLimitErrorRetries": 5}}}. IF null or {} then no router settings. - access_group_ids: Optional[List[str]] - List of access group IDs to associate with the key. Access groups define which models a key can access. Example - ["access_group_1", "access_group_2"]. - budget_limits: Optional[list] - List of concurrent budget windows for the key. Each window specifies a budget_limit, time_period, and optional budget_duration. Example - [{"budget_limit": 10.0, "time_period": "1d"}, {"budget_limit": 50.0, "time_period": "7d"}]. Example: ```bash curl --location 'http://0.0.0.0:4000/key/update' --header 'Authorization: Bearer sk-1234' --header 'Content-Type: application/json' --data '{ "key": "sk-1234", "key_alias": "my-key", "user_id": "user-1234", "team_id": "team-1234", "max_budget": 100, "metadata": {"any_key": "any-val"}, }' ``` */
+/** Update Key Fn Update an existing API key's parameters. The body is a merge patch: a field left out keeps its stored value, and on the key's own columns an explicit null clears it. The metadata-backed fields below are the exception, merging into the stored metadata instead: passing one as null leaves it unchanged, while `metadata` itself replaces the stored metadata wholesale. Parameters: - key: Optional[str] - The key to update. Either key or key_alias must be provided. - key_alias: Optional[str] - User-friendly key alias. If key is omitted, also identifies the key to update (must match exactly one key, same as /key/delete's key_aliases) - user_id: Optional[str] - User ID associated with key - team_id: Optional[str] - Team ID associated with key - agent_id: Optional[str] - The agent id associated with the key. - project_id: Optional[str] - Omit to retain the project, or send null to detach. A different project ID is rejected. - organization_id: Optional[str] - The organization id of the key. - budget_id: Optional[str] - The budget id associated with the key. Created by calling `/budget/new`. - end_user_budget_id: Optional[str] - Proxy admin only. Budget id applied to end users first seen through this key that carry no budget of their own. Omit to keep the current value, pass an empty string to clear it. - models: Optional[list] - Model_name's a user is allowed to call - tags: Optional[List[str]] - Tags for organizing keys (Enterprise only) - prompts: Optional[List[str]] - List of prompts that the key is allowed to use. - enforced_params: Optional[List[str]] - List of enforced params for the key (Enterprise only). [Docs](https://docs.litellm.ai/docs/proxy/enterprise#enforce-required-params-for-llm-requests) - spend: Optional[float] - Amount spent by key - max_budget: Optional[float] - Max budget for key - model_max_budget: Optional[Dict[str, BudgetConfig]] - Model-specific budgets {"gpt-4": {"budget_limit": 0.0005, "time_period": "30d"}} - budget_fallbacks: Optional[Dict[str, List[str]]] - Per-model fallback chain tried in order when that model's own `model_max_budget` is exceeded, e.g. {"gpt-4o": ["gpt-4o-mini"]}. - budget_duration: Optional[str] - Budget reset period ("30d", "1h", etc.) - soft_budget: Optional[float] - Soft budget limit (warning vs. hard stop). Will trigger a slack alert when this soft budget is reached. Set to null to remove the soft budget. - max_parallel_requests: Optional[int] - Rate limit for parallel requests - metadata: Optional[dict] - Metadata for key. Example {"team": "core-infra", "app": "app2"} - tpm_limit: Optional[int] - Tokens per minute limit - rpm_limit: Optional[int] - Requests per minute limit - tpd_limit: Optional[int] - Tokens per day limit, charged by batch submissions instead of tpm_limit/rpm_limit - model_rpm_limit: Optional[dict] - Model-specific RPM limits {"gpt-4": 100, "claude-v1": 200} - mcp_rpm_limit: Optional[dict] - Per-MCP-server RPM limits, keyed by MCP server name {"github": 100, "slack": 200} - tag_rpm_limit: Optional[dict] - Per-request-tag RPM limits, keyed by request tag {"cell-1": 1000, "cell-2": 500}. Each tag gets an independent counter; absent tags fall back to the key-level rpm limit. - model_tpm_limit: Optional[dict] - Model-specific TPM limits {"gpt-4": 100000, "claude-v1": 200000} - default_estimated_output_tokens: Optional[int] - Proxy admin only. Expected output tokens reserved for TPM limiting when a request omits max_tokens. Positive integer. - default_estimated_output_tokens_per_model: Optional[dict] - Proxy admin only. Per-model override of the above {"gpt-4": 4096, "gpt-3.5-turbo": 1024} - tpm_limit_type: Optional[str] - TPM rate limit type - "best_effort_throughput", "guaranteed_throughput", or "dynamic" - rpm_limit_type: Optional[str] - RPM rate limit type - "best_effort_throughput", "guaranteed_throughput", or "dynamic" - allowed_cache_controls: Optional[list] - List of allowed cache control values - duration: Optional[str] - Key validity duration ("30d", "1h", etc.), null to never expire, or "-1" to never expire (deprecated, use null) - permissions: Optional[dict] - Key-specific permissions - send_invite_email: Optional[bool] - Send invite email to user_id - guardrails: Optional[List[str]] - List of active guardrails for the key - policies: Optional[List[str]] - List of policy names to apply to the key. Policies define guardrails, conditions, and inheritance rules. - disable_global_guardrails: Optional[bool] - Whether to disable global guardrails for the key. - throttle_on_budget_exceeded: Optional[bool] - When the key exceeds its max_budget, throttle its tpm/rpm to the global budget_exceeded_throttle_percentage instead of blocking the key entirely. - enable_prompt_caching: Optional[bool] - Auto-inject prompt caching breakpoints (Anthropic cache_control markers) on requests made with this key. Supported Claude models on Anthropic, Bedrock, Vertex AI, and Azure AI only. - prompts: Optional[List[str]] - List of prompts that the key is allowed to use. - blocked: Optional[bool] - Whether the key is blocked - aliases: Optional[dict] - Model aliases for the key - [Docs](https://litellm.vercel.app/docs/proxy/virtual_keys#model-aliases) - config: Optional[dict] - [DEPRECATED PARAM] Key-specific config. - temp_budget_increase: Optional[float] - Temporary budget increase for the key (Enterprise only). - temp_budget_expiry: Optional[str] - Expiry time for the temporary budget increase (Enterprise only). - allowed_routes: Optional[list] - List of allowed routes for the key. Store the actual route or store a wildcard pattern for a set of routes. Example - ["/chat/completions", "/embeddings", "/keys/*"] - allowed_passthrough_routes: Optional[list] - List of allowed pass through routes for the key. Store the actual route or store a wildcard pattern for a set of routes. Example - ["/my-custom-endpoint"]. Use this instead of allowed_routes, if you just want to specify which pass through routes the key can access, without specifying the routes. If allowed_routes is specified, allowed_passthrough_routes is ignored. - prompts: Optional[List[str]] - List of allowed prompts for the key. If specified, the key will only be able to use these specific prompts. - object_permission: Optional[LiteLLM_ObjectPermissionBase] - key-specific object permission. Example - {"vector_stores": ["vector_store_1", "vector_store_2"], "agents": ["agent_1", "agent_2"], "agent_access_groups": ["dev_group"]}. IF null or {} then no object permission. - auto_rotate: Optional[bool] - Whether this key should be automatically rotated - rotation_interval: Optional[str] - How often to rotate this key (e.g., '30d', '90d'). Required if auto_rotate=True - allowed_vector_store_indexes: Optional[List[dict]] - List of allowed vector store indexes for the key. Example - [{"index_name": "my-index", "index_permissions": ["write", "read"]}]. If specified, the key will only be able to use these specific vector store indexes. Create index, using `/v1/indexes` endpoint. - router_settings: Optional[UpdateRouterConfig] - key-specific router settings. Example - {"model_group_retry_policy": {"gpt-4": {"RateLimitErrorRetries": 5}}}. IF null or {} then no router settings. - access_group_ids: Optional[List[str]] - List of access group IDs to associate with the key. Access groups define which models a key can access. Example - ["access_group_1", "access_group_2"]. - budget_limits: Optional[list] - List of concurrent budget windows for the key. Each window specifies a budget_limit, time_period, and optional budget_duration. Example - [{"budget_limit": 10.0, "time_period": "1d"}, {"budget_limit": 50.0, "time_period": "7d"}]. Example: ```bash curl --location 'http://0.0.0.0:4000/key/update' --header 'Authorization: Bearer sk-1234' --header 'Content-Type: application/json' --data '{ "key": "sk-1234", "key_alias": "my-key", "user_id": "user-1234", "team_id": "team-1234", "max_budget": 100, "metadata": {"any_key": "any-val"}, }' ``` */
 export const updateKeyFnKeyUpdatePost: API.OperationMethod<
   UpdateKeyFnKeyUpdatePostRequest,
   UpdateKeyFnKeyUpdatePostResponse,