@deeprelay/sdk 0.2.1 → 0.3.0

package/src/models/ChatCompletionRequest.ts

@@ -2,7 +2,7 @@
 /* eslint-disable */
 /**
  * deeprelay Public Developer API
- * Programmatic API for deeprelay. Authenticate with API keys minted from the dashboard. See https://deeprelay.ai/docs for details. 
+ * Programmatic API for deeprelay. Authenticate with API keys minted from the dashboard. See https://deeprelay.ai/docs for details.  Keys minted from the dashboard or `deeprelay login` belong to the Production server, https://api.deeprelay.ai/v1. The Demo server is an internal staging environment with its own key store.  Versioning: `/v1` is the API version. `info.version` is the revision of this `/v1` document, not a separate API.  OpenAI compatibility: chat completions follow the OpenAI request and response shapes, but not every OpenAI parameter is carried. Some are rejected with 400 `unsupported_parameter` and any parameter not declared in this spec is accepted and ignored. The full table is at https://deeprelay.ai/docs/api/inference#openai-compatibility. 
  *
  * The version of the OpenAPI document: 1.0.0
  * Contact: support@deeprelay.ai
@@ -13,6 +13,13 @@
  */
 
 import { mapValues } from '../runtime';
+import type { ChatCompletionRequestToolChoice } from './ChatCompletionRequestToolChoice';
+import {
+    ChatCompletionRequestToolChoiceFromJSON,
+    ChatCompletionRequestToolChoiceFromJSONTyped,
+    ChatCompletionRequestToolChoiceToJSON,
+    ChatCompletionRequestToolChoiceToJSONTyped,
+} from './ChatCompletionRequestToolChoice';
 import type { ChatMessage } from './ChatMessage';
 import {
     ChatMessageFromJSON,
@@ -27,6 +34,20 @@ import {
     StreamOptionsToJSON,
     StreamOptionsToJSONTyped,
 } from './StreamOptions';
+import type { ChatCompletionRequestResponseFormat } from './ChatCompletionRequestResponseFormat';
+import {
+    ChatCompletionRequestResponseFormatFromJSON,
+    ChatCompletionRequestResponseFormatFromJSONTyped,
+    ChatCompletionRequestResponseFormatToJSON,
+    ChatCompletionRequestResponseFormatToJSONTyped,
+} from './ChatCompletionRequestResponseFormat';
+import type { Tool } from './Tool';
+import {
+    ToolFromJSON,
+    ToolFromJSONTyped,
+    ToolToJSON,
+    ToolToJSONTyped,
+} from './Tool';
 
 /**
  * 
@@ -35,7 +56,7 @@ import {
  */
 export interface ChatCompletionRequest {
     /**
-     * Canonical model id, e.g. "deeprelay/qwen2.5-7b-instruct".
+     * Canonical model id, e.g. "deeprelay/llama-3.3-70b-instruct".
      * @type {string}
      * @memberof ChatCompletionRequest
      */
@@ -58,6 +79,13 @@ export interface ChatCompletionRequest {
      * @memberof ChatCompletionRequest
      */
     maxTokens?: number;
+    /**
+     * OpenAI's newer name for max_tokens, accepted for client compatibility. When both are sent, max_tokens wins.
+     * 
+     * @type {number}
+     * @memberof ChatCompletionRequest
+     */
+    maxCompletionTokens?: number;
     /**
      * 
      * @type {boolean}
@@ -70,6 +98,72 @@ export interface ChatCompletionRequest {
      * @memberof ChatCompletionRequest
      */
     streamOptions?: StreamOptions;
+    /**
+     * Functions the model may call. Support is per-model — check for "tools" in the model's supported_parameters before sending it. Omit the field entirely rather than sending an empty array.
+     * 
+     * @type {Array<Tool>}
+     * @memberof ChatCompletionRequest
+     */
+    tools?: Array<Tool>;
+    /**
+     * 
+     * @type {ChatCompletionRequestToolChoice}
+     * @memberof ChatCompletionRequest
+     */
+    toolChoice?: ChatCompletionRequestToolChoice;
+    /**
+     * 
+     * @type {ChatCompletionRequestResponseFormat}
+     * @memberof ChatCompletionRequest
+     */
+    responseFormat?: ChatCompletionRequestResponseFormat;
+    /**
+     * Hint: how much a reasoning model should think. Values in use are none, minimal, low, medium, high and xhigh; which ones a model accepts varies. Forwarded to an upstream that accepts the value, otherwise dropped and reported in x-deeprelay-ignored-params. Never rejected.
+     * 
+     * @type {string}
+     * @memberof ChatCompletionRequest
+     */
+    reasoningEffort?: string;
+    /**
+     * Hint. Forwarded when the serving upstream accepts it, otherwise dropped and reported in x-deeprelay-ignored-params.
+     * @type {number}
+     * @memberof ChatCompletionRequest
+     */
+    presencePenalty?: number;
+    /**
+     * Hint. Forwarded when the serving upstream accepts it, otherwise dropped and reported in x-deeprelay-ignored-params.
+     * @type {number}
+     * @memberof ChatCompletionRequest
+     */
+    frequencyPenalty?: number;
+    /**
+     * Hint for best-effort determinism. Forwarded when the serving upstream accepts it, otherwise dropped and reported. Identical seeds are not guaranteed to give identical output across upstreams.
+     * 
+     * @type {number}
+     * @memberof ChatCompletionRequest
+     */
+    seed?: number;
+    /**
+     * Whether the model may emit several tool calls in one turn. When false and the serving upstream cannot honour it, the relay returns only the first tool call and reports `parallel_tool_calls:enforced` in x-deeprelay-ignored-params.
+     * 
+     * @type {boolean}
+     * @memberof ChatCompletionRequest
+     */
+    parallelToolCalls?: boolean;
+    /**
+     * Return per-token log probabilities in choices[].logprobs. Gated per model ("logprobs" in supported_parameters). Values come from the serving upstream and are not comparable across upstreams or numeric precisions.
+     * 
+     * @type {boolean}
+     * @memberof ChatCompletionRequest
+     */
+    logprobs?: boolean;
+    /**
+     * Number of alternatives per token (requires logprobs). Gated per model; each model has its own maximum, and a larger value returns 400 unsupported_parameter.
+     * 
+     * @type {number}
+     * @memberof ChatCompletionRequest
+     */
+    topLogprobs?: number;
 }
 
 /**
@@ -95,8 +189,19 @@ export function ChatCompletionRequestFromJSONTyped(json: any, ignoreDiscriminato
         'messages': ((json['messages'] as Array<any>).map(ChatMessageFromJSON)),
         'temperature': json['temperature'] == null ? undefined : json['temperature'],
         'maxTokens': json['max_tokens'] == null ? undefined : json['max_tokens'],
+        'maxCompletionTokens': json['max_completion_tokens'] == null ? undefined : json['max_completion_tokens'],
         'stream': json['stream'] == null ? undefined : json['stream'],
         'streamOptions': json['stream_options'] == null ? undefined : StreamOptionsFromJSON(json['stream_options']),
+        'tools': json['tools'] == null ? undefined : ((json['tools'] as Array<any>).map(ToolFromJSON)),
+        'toolChoice': json['tool_choice'] == null ? undefined : ChatCompletionRequestToolChoiceFromJSON(json['tool_choice']),
+        'responseFormat': json['response_format'] == null ? undefined : ChatCompletionRequestResponseFormatFromJSON(json['response_format']),
+        'reasoningEffort': json['reasoning_effort'] == null ? undefined : json['reasoning_effort'],
+        'presencePenalty': json['presence_penalty'] == null ? undefined : json['presence_penalty'],
+        'frequencyPenalty': json['frequency_penalty'] == null ? undefined : json['frequency_penalty'],
+        'seed': json['seed'] == null ? undefined : json['seed'],
+        'parallelToolCalls': json['parallel_tool_calls'] == null ? undefined : json['parallel_tool_calls'],
+        'logprobs': json['logprobs'] == null ? undefined : json['logprobs'],
+        'topLogprobs': json['top_logprobs'] == null ? undefined : json['top_logprobs'],
     };
 }
 
@@ -115,8 +220,19 @@ export function ChatCompletionRequestToJSONTyped(value?: ChatCompletionRequest |
         'messages': ((value['messages'] as Array<any>).map(ChatMessageToJSON)),
         'temperature': value['temperature'],
         'max_tokens': value['maxTokens'],
+        'max_completion_tokens': value['maxCompletionTokens'],
         'stream': value['stream'],
         'stream_options': StreamOptionsToJSON(value['streamOptions']),
+        'tools': value['tools'] == null ? undefined : ((value['tools'] as Array<any>).map(ToolToJSON)),
+        'tool_choice': ChatCompletionRequestToolChoiceToJSON(value['toolChoice']),
+        'response_format': ChatCompletionRequestResponseFormatToJSON(value['responseFormat']),
+        'reasoning_effort': value['reasoningEffort'],
+        'presence_penalty': value['presencePenalty'],
+        'frequency_penalty': value['frequencyPenalty'],
+        'seed': value['seed'],
+        'parallel_tool_calls': value['parallelToolCalls'],
+        'logprobs': value['logprobs'],
+        'top_logprobs': value['topLogprobs'],
     };
 }
 

package/src/models/ChatCompletionRequestResponseFormat.ts

@@ -0,0 +1,84 @@
+/* tslint:disable */
+/* eslint-disable */
+/**
+ * deeprelay Public Developer API
+ * Programmatic API for deeprelay. Authenticate with API keys minted from the dashboard. See https://deeprelay.ai/docs for details.  Keys minted from the dashboard or `deeprelay login` belong to the Production server, https://api.deeprelay.ai/v1. The Demo server is an internal staging environment with its own key store.  Versioning: `/v1` is the API version. `info.version` is the revision of this `/v1` document, not a separate API.  OpenAI compatibility: chat completions follow the OpenAI request and response shapes, but not every OpenAI parameter is carried. Some are rejected with 400 `unsupported_parameter` and any parameter not declared in this spec is accepted and ignored. The full table is at https://deeprelay.ai/docs/api/inference#openai-compatibility. 
+ *
+ * The version of the OpenAPI document: 1.0.0
+ * Contact: support@deeprelay.ai
+ *
+ * NOTE: This class is auto generated by OpenAPI Generator (https://openapi-generator.tech).
+ * https://openapi-generator.tech
+ * Do not edit the class manually.
+ */
+
+import type { ResponseFormatJSONObject } from './ResponseFormatJSONObject';
+import {
+    instanceOfResponseFormatJSONObject,
+    ResponseFormatJSONObjectFromJSON,
+    ResponseFormatJSONObjectFromJSONTyped,
+    ResponseFormatJSONObjectToJSON,
+} from './ResponseFormatJSONObject';
+import type { ResponseFormatJSONSchema } from './ResponseFormatJSONSchema';
+import {
+    instanceOfResponseFormatJSONSchema,
+    ResponseFormatJSONSchemaFromJSON,
+    ResponseFormatJSONSchemaFromJSONTyped,
+    ResponseFormatJSONSchemaToJSON,
+} from './ResponseFormatJSONSchema';
+import type { ResponseFormatText } from './ResponseFormatText';
+import {
+    instanceOfResponseFormatText,
+    ResponseFormatTextFromJSON,
+    ResponseFormatTextFromJSONTyped,
+    ResponseFormatTextToJSON,
+} from './ResponseFormatText';
+
+/**
+ * @type ChatCompletionRequestResponseFormat
+ * Constrain the reply format. Gated per model: `json_object` needs "response_format" in the model's supported_parameters, and a strict `json_schema` needs "structured_outputs". A model without the capability returns 400 unsupported_parameter naming the models that have it. A non-strict json_schema on a model with only "response_format" is served as json_object and reported as `response_format:downgraded` in x-deeprelay-ignored-params. The relay never emulates JSON mode by prompting.
+ * 
+ * @export
+ */
+export type ChatCompletionRequestResponseFormat = { type: 'json_object' } & ResponseFormatJSONObject | { type: 'json_schema' } & ResponseFormatJSONSchema | { type: 'text' } & ResponseFormatText;
+
+export function ChatCompletionRequestResponseFormatFromJSON(json: any): ChatCompletionRequestResponseFormat {
+    return ChatCompletionRequestResponseFormatFromJSONTyped(json, false);
+}
+
+export function ChatCompletionRequestResponseFormatFromJSONTyped(json: any, ignoreDiscriminator: boolean): ChatCompletionRequestResponseFormat {
+    if (json == null) {
+        return json;
+    }
+    switch (json['type']) {
+        case 'json_object':
+            return Object.assign({}, ResponseFormatJSONObjectFromJSONTyped(json, true), { type: 'json_object' } as const);
+        case 'json_schema':
+            return Object.assign({}, ResponseFormatJSONSchemaFromJSONTyped(json, true), { type: 'json_schema' } as const);
+        case 'text':
+            return Object.assign({}, ResponseFormatTextFromJSONTyped(json, true), { type: 'text' } as const);
+        default:
+            return json;
+    }
+}
+
+export function ChatCompletionRequestResponseFormatToJSON(json: any): any {
+    return ChatCompletionRequestResponseFormatToJSONTyped(json, false);
+}
+
+export function ChatCompletionRequestResponseFormatToJSONTyped(value?: ChatCompletionRequestResponseFormat | null, ignoreDiscriminator: boolean = false): any {
+    if (value == null) {
+        return value;
+    }
+    switch (value['type']) {
+        case 'json_object':
+            return Object.assign({}, ResponseFormatJSONObjectToJSON(value), { 'type': 'json_object' } as const);
+        case 'json_schema':
+            return Object.assign({}, ResponseFormatJSONSchemaToJSON(value), { 'type': 'json_schema' } as const);
+        case 'text':
+            return Object.assign({}, ResponseFormatTextToJSON(value), { 'type': 'text' } as const);
+        default:
+            return value;
+    }
+}
+

package/src/models/ChatCompletionRequestToolChoice.ts

@@ -0,0 +1,70 @@
+/* tslint:disable */
+/* eslint-disable */
+/**
+ * deeprelay Public Developer API
+ * Programmatic API for deeprelay. Authenticate with API keys minted from the dashboard. See https://deeprelay.ai/docs for details.  Keys minted from the dashboard or `deeprelay login` belong to the Production server, https://api.deeprelay.ai/v1. The Demo server is an internal staging environment with its own key store.  Versioning: `/v1` is the API version. `info.version` is the revision of this `/v1` document, not a separate API.  OpenAI compatibility: chat completions follow the OpenAI request and response shapes, but not every OpenAI parameter is carried. Some are rejected with 400 `unsupported_parameter` and any parameter not declared in this spec is accepted and ignored. The full table is at https://deeprelay.ai/docs/api/inference#openai-compatibility. 
+ *
+ * The version of the OpenAPI document: 1.0.0
+ * Contact: support@deeprelay.ai
+ *
+ * NOTE: This class is auto generated by OpenAPI Generator (https://openapi-generator.tech).
+ * https://openapi-generator.tech
+ * Do not edit the class manually.
+ */
+
+import type { ToolChoiceFunction } from './ToolChoiceFunction';
+import {
+    instanceOfToolChoiceFunction,
+    ToolChoiceFunctionFromJSON,
+    ToolChoiceFunctionFromJSONTyped,
+    ToolChoiceFunctionToJSON,
+} from './ToolChoiceFunction';
+
+/**
+ * @type ChatCompletionRequestToolChoice
+ * How the model should pick a tool: the strings "none", "auto" or "required", or an object naming one function. Requires a non-empty tools array unless it is "none".
+ * 
+ * @export
+ */
+export type ChatCompletionRequestToolChoice = ToolChoiceFunction | string;
+
+export function ChatCompletionRequestToolChoiceFromJSON(json: any): ChatCompletionRequestToolChoice {
+    return ChatCompletionRequestToolChoiceFromJSONTyped(json, false);
+}
+
+export function ChatCompletionRequestToolChoiceFromJSONTyped(json: any, ignoreDiscriminator: boolean): ChatCompletionRequestToolChoice {
+    if (json == null) {
+        return json;
+    }
+    if (typeof json !== 'object') {
+        return json;
+    }
+    if (instanceOfToolChoiceFunction(json)) {
+        return ToolChoiceFunctionFromJSONTyped(json, true);
+    }
+    if (typeof json === 'string' && (json === 'none' || json === 'auto' || json === 'required')) {
+        return json;
+    }
+    return {} as any;
+}
+
+export function ChatCompletionRequestToolChoiceToJSON(json: any): any {
+    return ChatCompletionRequestToolChoiceToJSONTyped(json, false);
+}
+
+export function ChatCompletionRequestToolChoiceToJSONTyped(value?: ChatCompletionRequestToolChoice | null, ignoreDiscriminator: boolean = false): any {
+    if (value == null) {
+        return value;
+    }
+    if (typeof value !== 'object') {
+        return value;
+    }
+    if (instanceOfToolChoiceFunction(value)) {
+        return ToolChoiceFunctionToJSON(value as ToolChoiceFunction);
+    }
+    if (typeof value === 'string' && (value === 'none' || value === 'auto' || value === 'required')) {
+        return value;
+    }
+    return {};
+}
+

package/src/models/ChatCompletionResponse.ts

@@ -2,7 +2,7 @@
 /* eslint-disable */
 /**
  * deeprelay Public Developer API
- * Programmatic API for deeprelay. Authenticate with API keys minted from the dashboard. See https://deeprelay.ai/docs for details. 
+ * Programmatic API for deeprelay. Authenticate with API keys minted from the dashboard. See https://deeprelay.ai/docs for details.  Keys minted from the dashboard or `deeprelay login` belong to the Production server, https://api.deeprelay.ai/v1. The Demo server is an internal staging environment with its own key store.  Versioning: `/v1` is the API version. `info.version` is the revision of this `/v1` document, not a separate API.  OpenAI compatibility: chat completions follow the OpenAI request and response shapes, but not every OpenAI parameter is carried. Some are rejected with 400 `unsupported_parameter` and any parameter not declared in this spec is accepted and ignored. The full table is at https://deeprelay.ai/docs/api/inference#openai-compatibility. 
  *
  * The version of the OpenAPI document: 1.0.0
  * Contact: support@deeprelay.ai

package/src/models/ChatMessage.ts

@@ -2,7 +2,7 @@
 /* eslint-disable */
 /**
  * deeprelay Public Developer API
- * Programmatic API for deeprelay. Authenticate with API keys minted from the dashboard. See https://deeprelay.ai/docs for details. 
+ * Programmatic API for deeprelay. Authenticate with API keys minted from the dashboard. See https://deeprelay.ai/docs for details.  Keys minted from the dashboard or `deeprelay login` belong to the Production server, https://api.deeprelay.ai/v1. The Demo server is an internal staging environment with its own key store.  Versioning: `/v1` is the API version. `info.version` is the revision of this `/v1` document, not a separate API.  OpenAI compatibility: chat completions follow the OpenAI request and response shapes, but not every OpenAI parameter is carried. Some are rejected with 400 `unsupported_parameter` and any parameter not declared in this spec is accepted and ignored. The full table is at https://deeprelay.ai/docs/api/inference#openai-compatibility. 
  *
  * The version of the OpenAPI document: 1.0.0
  * Contact: support@deeprelay.ai
@@ -13,6 +13,14 @@
  */
 
 import { mapValues } from '../runtime';
+import type { ToolCall } from './ToolCall';
+import {
+    ToolCallFromJSON,
+    ToolCallFromJSONTyped,
+    ToolCallToJSON,
+    ToolCallToJSONTyped,
+} from './ToolCall';
+
 /**
  * 
  * @export
@@ -20,17 +28,34 @@ import { mapValues } from '../runtime';
  */
 export interface ChatMessage {
     /**
+     * "developer" is OpenAI's newer spelling of "system" and is treated as system.
      * 
      * @type {ChatMessageRoleEnum}
      * @memberof ChatMessage
      */
     role: ChatMessageRoleEnum;
     /**
+     * The message text. Null on an assistant turn that is purely a tool call — branch on that to tell a tool call from a reply. Responses always carry a string (or null).
+     * 
+     * Requests may instead send an array of content parts, as OpenAI clients do (typed here as a string so SDK callers can keep passing plain text). Text parts ({"type":"text","text":"..."}) are joined with newlines. Image parts ({"type":"image_url","image_url":{"url":"..."}}) are accepted on models whose supported_parameters list "vision"; any other model returns 400 unsupported_parameter (param messages) naming the models that accept images. `url` must be an https URL or a data:image/{png,jpeg,webp,gif};base64 URI, and the relay never fetches it itself. Limits: 16 images per request, 20 MiB per image, 8 KiB per URL, inside the 32 MiB body cap (413 request_too_large). Each image counts as 1,600 tokens in the spend preflight. `image_url.detail` is accepted but not forwarded, and is reported in x-deeprelay-ignored-params. Audio and file parts are rejected with 400 unsupported_content_part.
+     * 
+     * @type {string}
+     * @memberof ChatMessage
+     */
+    content?: string;
+    /**
+     * Calls the model wants made. Assistant messages only.
+     * @type {Array<ToolCall>}
+     * @memberof ChatMessage
+     */
+    toolCalls?: Array<ToolCall>;
+    /**
+     * Which call this message answers. REQUIRED on a role "tool" message — without it the upstream cannot match the result to its call.
      * 
      * @type {string}
      * @memberof ChatMessage
      */
-    content: string;
+    toolCallId?: string;
 }
 
 
@@ -39,8 +64,10 @@ export interface ChatMessage {
  */
 export const ChatMessageRoleEnum = {
     System: 'system',
+    Developer: 'developer',
     User: 'user',
-    Assistant: 'assistant'
+    Assistant: 'assistant',
+    Tool: 'tool'
 } as const;
 export type ChatMessageRoleEnum = typeof ChatMessageRoleEnum[keyof typeof ChatMessageRoleEnum];
 
@@ -50,7 +77,6 @@ export type ChatMessageRoleEnum = typeof ChatMessageRoleEnum[keyof typeof ChatMe
  */
 export function instanceOfChatMessage(value: object): value is ChatMessage {
     if (!('role' in value) || value['role'] === undefined) return false;
-    if (!('content' in value) || value['content'] === undefined) return false;
     return true;
 }
 
@@ -65,7 +91,9 @@ export function ChatMessageFromJSONTyped(json: any, ignoreDiscriminator: boolean
     return {
         
         'role': json['role'],
-        'content': json['content'],
+        'content': json['content'] == null ? undefined : json['content'],
+        'toolCalls': json['tool_calls'] == null ? undefined : ((json['tool_calls'] as Array<any>).map(ToolCallFromJSON)),
+        'toolCallId': json['tool_call_id'] == null ? undefined : json['tool_call_id'],
     };
 }
 
@@ -82,6 +110,8 @@ export function ChatMessageToJSONTyped(value?: ChatMessage | null, ignoreDiscrim
         
         'role': value['role'],
         'content': value['content'],
+        'tool_calls': value['toolCalls'] == null ? undefined : ((value['toolCalls'] as Array<any>).map(ToolCallToJSON)),
+        'tool_call_id': value['toolCallId'],
     };
 }
 

package/src/models/CostDetails.ts

@@ -0,0 +1,105 @@
+/* tslint:disable */
+/* eslint-disable */
+/**
+ * deeprelay Public Developer API
+ * Programmatic API for deeprelay. Authenticate with API keys minted from the dashboard. See https://deeprelay.ai/docs for details.  Keys minted from the dashboard or `deeprelay login` belong to the Production server, https://api.deeprelay.ai/v1. The Demo server is an internal staging environment with its own key store.  Versioning: `/v1` is the API version. `info.version` is the revision of this `/v1` document, not a separate API.  OpenAI compatibility: chat completions follow the OpenAI request and response shapes, but not every OpenAI parameter is carried. Some are rejected with 400 `unsupported_parameter` and any parameter not declared in this spec is accepted and ignored. The full table is at https://deeprelay.ai/docs/api/inference#openai-compatibility. 
+ *
+ * The version of the OpenAPI document: 1.0.0
+ * Contact: support@deeprelay.ai
+ *
+ * NOTE: This class is auto generated by OpenAPI Generator (https://openapi-generator.tech).
+ * https://openapi-generator.tech
+ * Do not edit the class manually.
+ */
+
+import { mapValues } from '../runtime';
+/**
+ * How this call was priced, in USD (exact decimals, up to 8 places). list_price_equivalent = plan_covered_amount + charged_amount. A call is either fully covered by the subscription or fully charged: there is no partial coverage.
+ * 
+ * @export
+ * @interface CostDetails
+ */
+export interface CostDetails {
+    /**
+     * What the call costs at the model's pay-as-you-go list rates (peak rates inside a peak window).
+     * @type {number}
+     * @memberof CostDetails
+     */
+    listPriceEquivalent: number;
+    /**
+     * The part the flat subscription covered (the full list price, or 0).
+     * @type {number}
+     * @memberof CostDetails
+     */
+    planCoveredAmount: number;
+    /**
+     * The part charged to the wallet; equal to usage.cost.
+     * @type {number}
+     * @memberof CostDetails
+     */
+    chargedAmount: number;
+    /**
+     * Weighted subscription-quota units this call consumed: (prompt + completion tokens) × the model's usage factor. 0 for a subscriber call billed pay-as-you-go (a model the plan does not cover, or a quota fallback); null for a pay-as-you-go organization.
+     * 
+     * @type {number}
+     * @memberof CostDetails
+     */
+    quotaUnitsUsed: number | null;
+    /**
+     * What the prompt cache saved: the list price without the cache discount minus list_price_equivalent. Null when the upstream did not report cache usage. Can be negative when a model's cached rate is not a discount.
+     * 
+     * @type {number}
+     * @memberof CostDetails
+     */
+    cacheSavings: number | null;
+}
+
+/**
+ * Check if a given object implements the CostDetails interface.
+ */
+export function instanceOfCostDetails(value: object): value is CostDetails {
+    if ((!('listPriceEquivalent' in (value as Record<string, any>)) && !('list_price_equivalent' in (value as Record<string, any>))) || ((value as Record<string, any>)['listPriceEquivalent'] === undefined && (value as Record<string, any>)['list_price_equivalent'] === undefined)) return false;
+    if ((!('planCoveredAmount' in (value as Record<string, any>)) && !('plan_covered_amount' in (value as Record<string, any>))) || ((value as Record<string, any>)['planCoveredAmount'] === undefined && (value as Record<string, any>)['plan_covered_amount'] === undefined)) return false;
+    if ((!('chargedAmount' in (value as Record<string, any>)) && !('charged_amount' in (value as Record<string, any>))) || ((value as Record<string, any>)['chargedAmount'] === undefined && (value as Record<string, any>)['charged_amount'] === undefined)) return false;
+    if ((!('quotaUnitsUsed' in (value as Record<string, any>)) && !('quota_units_used' in (value as Record<string, any>))) || ((value as Record<string, any>)['quotaUnitsUsed'] === undefined && (value as Record<string, any>)['quota_units_used'] === undefined)) return false;
+    if ((!('cacheSavings' in (value as Record<string, any>)) && !('cache_savings' in (value as Record<string, any>))) || ((value as Record<string, any>)['cacheSavings'] === undefined && (value as Record<string, any>)['cache_savings'] === undefined)) return false;
+    return true;
+}
+
+export function CostDetailsFromJSON(json: any): CostDetails {
+    return CostDetailsFromJSONTyped(json, false);
+}
+
+export function CostDetailsFromJSONTyped(json: any, ignoreDiscriminator: boolean): CostDetails {
+    if (json == null) {
+        return json;
+    }
+    return {
+        
+        'listPriceEquivalent': json['list_price_equivalent'],
+        'planCoveredAmount': json['plan_covered_amount'],
+        'chargedAmount': json['charged_amount'],
+        'quotaUnitsUsed': json['quota_units_used'],
+        'cacheSavings': json['cache_savings'],
+    };
+}
+
+export function CostDetailsToJSON(json: any): CostDetails {
+    return CostDetailsToJSONTyped(json, false);
+}
+
+export function CostDetailsToJSONTyped(value?: CostDetails | null, ignoreDiscriminator: boolean = false): any {
+    if (value == null) {
+        return value;
+    }
+
+    return {
+        
+        'list_price_equivalent': value['listPriceEquivalent'],
+        'plan_covered_amount': value['planCoveredAmount'],
+        'charged_amount': value['chargedAmount'],
+        'quota_units_used': value['quotaUnitsUsed'],
+        'cache_savings': value['cacheSavings'],
+    };
+}
+

package/src/models/CreateCryptoDepositRequest.ts

@@ -2,7 +2,7 @@
 /* eslint-disable */
 /**
  * deeprelay Public Developer API
- * Programmatic API for deeprelay. Authenticate with API keys minted from the dashboard. See https://deeprelay.ai/docs for details. 
+ * Programmatic API for deeprelay. Authenticate with API keys minted from the dashboard. See https://deeprelay.ai/docs for details.  Keys minted from the dashboard or `deeprelay login` belong to the Production server, https://api.deeprelay.ai/v1. The Demo server is an internal staging environment with its own key store.  Versioning: `/v1` is the API version. `info.version` is the revision of this `/v1` document, not a separate API.  OpenAI compatibility: chat completions follow the OpenAI request and response shapes, but not every OpenAI parameter is carried. Some are rejected with 400 `unsupported_parameter` and any parameter not declared in this spec is accepted and ignored. The full table is at https://deeprelay.ai/docs/api/inference#openai-compatibility. 
  *
  * The version of the OpenAPI document: 1.0.0
  * Contact: support@deeprelay.ai
@@ -27,6 +27,7 @@ export interface CreateCryptoDepositRequest {
      */
     amountCents: number;
     /**
+     * Network to send on. tron and bsc (BNB Chain) carry USDT only (TRC-20 and BEP-20); base carries USDC only.
      * 
      * @type {CreateCryptoDepositRequestChainEnum}
      * @memberof CreateCryptoDepositRequest
@@ -49,7 +50,9 @@ export const CreateCryptoDepositRequestChainEnum = {
     Base: 'base',
     Arbitrum: 'arbitrum',
     Polygon: 'polygon',
-    Solana: 'solana'
+    Solana: 'solana',
+    Tron: 'tron',
+    Bsc: 'bsc'
 } as const;
 export type CreateCryptoDepositRequestChainEnum = typeof CreateCryptoDepositRequestChainEnum[keyof typeof CreateCryptoDepositRequestChainEnum];
 

package/src/models/CreateWebhookEndpointRequest.ts

@@ -2,7 +2,7 @@
 /* eslint-disable */
 /**
  * deeprelay Public Developer API
- * Programmatic API for deeprelay. Authenticate with API keys minted from the dashboard. See https://deeprelay.ai/docs for details. 
+ * Programmatic API for deeprelay. Authenticate with API keys minted from the dashboard. See https://deeprelay.ai/docs for details.  Keys minted from the dashboard or `deeprelay login` belong to the Production server, https://api.deeprelay.ai/v1. The Demo server is an internal staging environment with its own key store.  Versioning: `/v1` is the API version. `info.version` is the revision of this `/v1` document, not a separate API.  OpenAI compatibility: chat completions follow the OpenAI request and response shapes, but not every OpenAI parameter is carried. Some are rejected with 400 `unsupported_parameter` and any parameter not declared in this spec is accepted and ignored. The full table is at https://deeprelay.ai/docs/api/inference#openai-compatibility. 
  *
  * The version of the OpenAPI document: 1.0.0
  * Contact: support@deeprelay.ai
@@ -43,7 +43,11 @@ export const CreateWebhookEndpointRequestEventTypesEnum = {
     InstanceTerminated: 'instance.terminated',
     InstanceFailed: 'instance.failed',
     VideoCompleted: 'video.completed',
-    VideoFailed: 'video.failed'
+    VideoFailed: 'video.failed',
+    BalanceLow: 'balance.low',
+    SubscriptionQuotaThreshold: 'subscription_quota.threshold',
+    ApiKeyBudgetThreshold: 'api_key_budget.threshold',
+    MemberSpendThreshold: 'member_spend.threshold'
 } as const;
 export type CreateWebhookEndpointRequestEventTypesEnum = typeof CreateWebhookEndpointRequestEventTypesEnum[keyof typeof CreateWebhookEndpointRequestEventTypesEnum];
 

package/src/models/CryptoDeposit.ts

@@ -2,7 +2,7 @@
 /* eslint-disable */
 /**
  * deeprelay Public Developer API
- * Programmatic API for deeprelay. Authenticate with API keys minted from the dashboard. See https://deeprelay.ai/docs for details. 
+ * Programmatic API for deeprelay. Authenticate with API keys minted from the dashboard. See https://deeprelay.ai/docs for details.  Keys minted from the dashboard or `deeprelay login` belong to the Production server, https://api.deeprelay.ai/v1. The Demo server is an internal staging environment with its own key store.  Versioning: `/v1` is the API version. `info.version` is the revision of this `/v1` document, not a separate API.  OpenAI compatibility: chat completions follow the OpenAI request and response shapes, but not every OpenAI parameter is carried. Some are rejected with 400 `unsupported_parameter` and any parameter not declared in this spec is accepted and ignored. The full table is at https://deeprelay.ai/docs/api/inference#openai-compatibility. 
  *
  * The version of the OpenAPI document: 1.0.0
  * Contact: support@deeprelay.ai
@@ -99,7 +99,9 @@ export const CryptoDepositChainEnum = {
     Base: 'base',
     Arbitrum: 'arbitrum',
     Polygon: 'polygon',
-    Solana: 'solana'
+    Solana: 'solana',
+    Tron: 'tron',
+    Bsc: 'bsc'
 } as const;
 export type CryptoDepositChainEnum = typeof CryptoDepositChainEnum[keyof typeof CryptoDepositChainEnum];
 

package/src/models/Embedding.ts

@@ -2,7 +2,7 @@
 /* eslint-disable */
 /**
  * deeprelay Public Developer API
- * Programmatic API for deeprelay. Authenticate with API keys minted from the dashboard. See https://deeprelay.ai/docs for details. 
+ * Programmatic API for deeprelay. Authenticate with API keys minted from the dashboard. See https://deeprelay.ai/docs for details.  Keys minted from the dashboard or `deeprelay login` belong to the Production server, https://api.deeprelay.ai/v1. The Demo server is an internal staging environment with its own key store.  Versioning: `/v1` is the API version. `info.version` is the revision of this `/v1` document, not a separate API.  OpenAI compatibility: chat completions follow the OpenAI request and response shapes, but not every OpenAI parameter is carried. Some are rejected with 400 `unsupported_parameter` and any parameter not declared in this spec is accepted and ignored. The full table is at https://deeprelay.ai/docs/api/inference#openai-compatibility. 
  *
  * The version of the OpenAPI document: 1.0.0
  * Contact: support@deeprelay.ai

package/src/models/EmbeddingsRequest.ts

@@ -2,7 +2,7 @@
 /* eslint-disable */
 /**
  * deeprelay Public Developer API
- * Programmatic API for deeprelay. Authenticate with API keys minted from the dashboard. See https://deeprelay.ai/docs for details. 
+ * Programmatic API for deeprelay. Authenticate with API keys minted from the dashboard. See https://deeprelay.ai/docs for details.  Keys minted from the dashboard or `deeprelay login` belong to the Production server, https://api.deeprelay.ai/v1. The Demo server is an internal staging environment with its own key store.  Versioning: `/v1` is the API version. `info.version` is the revision of this `/v1` document, not a separate API.  OpenAI compatibility: chat completions follow the OpenAI request and response shapes, but not every OpenAI parameter is carried. Some are rejected with 400 `unsupported_parameter` and any parameter not declared in this spec is accepted and ignored. The full table is at https://deeprelay.ai/docs/api/inference#openai-compatibility. 
  *
  * The version of the OpenAPI document: 1.0.0
  * Contact: support@deeprelay.ai

package/src/models/EmbeddingsRequestInput.ts

@@ -2,7 +2,7 @@
 /* eslint-disable */
 /**
  * deeprelay Public Developer API
- * Programmatic API for deeprelay. Authenticate with API keys minted from the dashboard. See https://deeprelay.ai/docs for details. 
+ * Programmatic API for deeprelay. Authenticate with API keys minted from the dashboard. See https://deeprelay.ai/docs for details.  Keys minted from the dashboard or `deeprelay login` belong to the Production server, https://api.deeprelay.ai/v1. The Demo server is an internal staging environment with its own key store.  Versioning: `/v1` is the API version. `info.version` is the revision of this `/v1` document, not a separate API.  OpenAI compatibility: chat completions follow the OpenAI request and response shapes, but not every OpenAI parameter is carried. Some are rejected with 400 `unsupported_parameter` and any parameter not declared in this spec is accepted and ignored. The full table is at https://deeprelay.ai/docs/api/inference#openai-compatibility. 
  *
  * The version of the OpenAPI document: 1.0.0
  * Contact: support@deeprelay.ai