@deeprelay/sdk 0.2.1 → 0.3.0

package/src/models/EmbeddingsResponse.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/EmbeddingsUsage.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/GetHealth200Response.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/ImagesGenerationsRequest.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
@@ -20,7 +20,7 @@ import { mapValues } from '../runtime';
  */
 export interface ImagesGenerationsRequest {
     /**
-     * Customer-facing image model id (e.g. deeprelay/flux.1-schnell) or alias.
+     * Customer-facing image model id (e.g. deeprelay/seedream-4.0) or alias.
      * @type {string}
      * @memberof ImagesGenerationsRequest
      */

package/src/models/ImagesResponse.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/ImagesResponseDataInner.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/ImagesResponseUsage.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/InferencePreflight.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
@@ -33,7 +33,7 @@ export interface InferencePreflight {
      */
     verdict: InferencePreflightVerdictEnum;
     /**
-     * Stable code explaining the verdict, for clients that need their own wording. Absent on a plain `ok` that needs no explanation.
+     * Stable code explaining the verdict, for clients that need their own wording. Absent on a plain `ok` that needs no explanation. `api_key_budget_exceeded`, `member_spend_limit_reached` and `model_not_allowed_for_key` come with `verdict: block` and only for a key that has a budget or a model allowlist, or a member that has a spend limit.
      * @type {InferencePreflightReasonEnum}
      * @memberof InferencePreflight
      */
@@ -86,6 +86,9 @@ export const InferencePreflightReasonEnum = {
     InsufficientBalance: 'insufficient_balance',
     DailyLimitReached: 'daily_limit_reached',
     MonthlyLimitReached: 'monthly_limit_reached',
+    ApiKeyBudgetExceeded: 'api_key_budget_exceeded',
+    MemberSpendLimitReached: 'member_spend_limit_reached',
+    ModelNotAllowedForKey: 'model_not_allowed_for_key',
     PreflightUnavailable: 'preflight_unavailable'
 } as const;
 export type InferencePreflightReasonEnum = typeof InferencePreflightReasonEnum[keyof typeof InferencePreflightReasonEnum];

package/src/models/JSONSchemaFormat.ts

@@ -0,0 +1,92 @@
+/* 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';
+/**
+ * The schema is forwarded byte-for-byte. It may be at most 64 KiB and nested at most 32 levels deep; larger or deeper schemas return 400 unsupported_parameter (param response_format).
+ * 
+ * @export
+ * @interface JSONSchemaFormat
+ */
+export interface JSONSchemaFormat {
+    /**
+     * 
+     * @type {string}
+     * @memberof JSONSchemaFormat
+     */
+    name: string;
+    /**
+     * 
+     * @type {string}
+     * @memberof JSONSchemaFormat
+     */
+    description?: string;
+    /**
+     * A JSON Schema object describing the reply.
+     * @type {{ [key: string]: any; }}
+     * @memberof JSONSchemaFormat
+     */
+    schema?: { [key: string]: any; };
+    /**
+     * When true, the reply is guaranteed to match the schema; only models listing "structured_outputs" accept it.
+     * 
+     * @type {boolean}
+     * @memberof JSONSchemaFormat
+     */
+    strict?: boolean;
+}
+
+/**
+ * Check if a given object implements the JSONSchemaFormat interface.
+ */
+export function instanceOfJSONSchemaFormat(value: object): value is JSONSchemaFormat {
+    if (!('name' in value) || value['name'] === undefined) return false;
+    return true;
+}
+
+export function JSONSchemaFormatFromJSON(json: any): JSONSchemaFormat {
+    return JSONSchemaFormatFromJSONTyped(json, false);
+}
+
+export function JSONSchemaFormatFromJSONTyped(json: any, ignoreDiscriminator: boolean): JSONSchemaFormat {
+    if (json == null) {
+        return json;
+    }
+    return {
+        
+        'name': json['name'],
+        'description': json['description'] == null ? undefined : json['description'],
+        'schema': json['schema'] == null ? undefined : json['schema'],
+        'strict': json['strict'] == null ? undefined : json['strict'],
+    };
+}
+
+export function JSONSchemaFormatToJSON(json: any): JSONSchemaFormat {
+    return JSONSchemaFormatToJSONTyped(json, false);
+}
+
+export function JSONSchemaFormatToJSONTyped(value?: JSONSchemaFormat | null, ignoreDiscriminator: boolean = false): any {
+    if (value == null) {
+        return value;
+    }
+
+    return {
+        
+        'name': value['name'],
+        'description': value['description'],
+        'schema': value['schema'],
+        'strict': value['strict'],
+    };
+}
+

package/src/models/ListDeposits200Response.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/Model.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 { ModelMetrics } from './ModelMetrics';
+import {
+    ModelMetricsFromJSON,
+    ModelMetricsFromJSONTyped,
+    ModelMetricsToJSON,
+    ModelMetricsToJSONTyped,
+} from './ModelMetrics';
 import type { ModelPricing } from './ModelPricing';
 import {
     ModelPricingFromJSON,
@@ -47,13 +54,13 @@ export interface Model {
      */
     object: ModelObjectEnum;
     /**
-     * 
+     * Always present for OpenAI-client compatibility. Currently 0 for every model (the catalog carries no creation date); do not sort on it.
      * @type {number}
      * @memberof Model
      */
-    created?: number;
+    created: number;
     /**
-     * 
+     * The model's author (the same value as `author`), or "deeprelay" when the author is unknown.
      * @type {string}
      * @memberof Model
      */
@@ -76,18 +83,36 @@ export interface Model {
      * @memberof Model
      */
     category?: string;
+    /**
+     * The serving tier this entry prices: `serverless` or `economy` today, and more values may be added. A model served on both tiers appears twice: under its bare id for `serverless`, and under the `:economy`-suffixed id for `economy`. Send the id of the tier you want. Omitted on `coming_soon` placeholder entries.
+     * @type {string}
+     * @memberof Model
+     */
+    tier?: string;
     /**
      * 
      * @type {number}
      * @memberof Model
      */
     contextLength?: number;
+    /**
+     * The model's maximum output, in tokens, from the catalog. Omitted when the catalog lists none.
+     * @type {number}
+     * @memberof Model
+     */
+    maxCompletionTokens?: number;
     /**
      * 
      * @type {Array<string>}
      * @memberof Model
      */
     supportedParameters?: Array<string>;
+    /**
+     * The `reasoning_effort` values this model accepts (for example `none`, `low`, `medium`, `high`), as the catalog lists them. Omitted when the catalog lists none.
+     * @type {Array<string>}
+     * @memberof Model
+     */
+    reasoningEfforts?: Array<string>;
     /**
      * Typed parameter descriptors for media (image/video) models — name, type, enum/bounds/default. Playgrounds render input forms from these.
      * @type {Array<ModelParametersInner>}
@@ -119,19 +144,49 @@ export interface Model {
      */
     fineTuneBaseModel?: string;
     /**
-     * Whether the flat subscription tier covers this model. A covered model rides the plan's quota while the subscription is entitled and has quota left; everything else bills pay-as-you-go at list rates, as does a covered model once the quota is spent. Absent means false.
+     * The Hugging Face repo that publishes this model's open weights, e.g. `meta-llama/Llama-3.3-70B-Instruct`. Send it as `base_model` to run the same model on a dedicated deployment. Absent when the weights are not public.
+     * @type {string}
+     * @memberof Model
+     */
+    hfRepo?: string;
+    /**
+     * How heavily this model draws on the subscription quota, ×1000: each token counts as usage_factor_milli / 1000 weighted tokens toward the quota (and in cost_details.quota_units_used). Omitted at the default 1000 (1.0×).
+     * @type {number}
+     * @memberof Model
+     */
+    usageFactorMilli?: number;
+    /**
+     * "Included in the flat subscription": whether the subscription tier covers this model. A model-level fact, the same for every caller. A covered model rides the plan's quota while the subscription is entitled and has quota left; everything else bills pay-as-you-go at list rates, as does a covered model once the quota is spent. Absent means false.
      * 
      * Coverage is a property of the platform, not of the caller: the flag reads the same for every organization, subscribed or not. Use `/billing/subscription` for what the caller's own plan has left.
      * @type {boolean}
      * @memberof Model
      */
     planCovered?: boolean;
+    /**
+     * The audiences deeprelay recommends this model for. The only value today is `coding-agents`: models tested with coding agents, which `deeprelay setup <agent>` configures. More values may be added. Omitted on every model that is not recommended.
+     * @type {Array<string>}
+     * @memberof Model
+     */
+    recommendedFor?: Array<string>;
+    /**
+     * The model's position in the coding-agent list, 1 first. Present exactly when `recommended_for` contains `coding-agents`; a setup command makes the rank-1 model the default.
+     * @type {number}
+     * @memberof Model
+     */
+    codingAgentRank?: number;
     /**
      * 
      * @type {ModelStatusEnum}
      * @memberof Model
      */
     status: ModelStatusEnum;
+    /**
+     * 
+     * @type {ModelMetrics}
+     * @memberof Model
+     */
+    metrics?: ModelMetrics;
 }
 
 
@@ -173,6 +228,7 @@ export function instanceOfModel(value: object): value is Model {
     if (!('object' in value) || value['object'] === undefined) return false;
     if (value['object'] !== 'model') return false;
     
+    if (!('created' in value) || value['created'] === undefined) return false;
     if ((!('ownedBy' in (value as Record<string, any>)) && !('owned_by' in (value as Record<string, any>))) || ((value as Record<string, any>)['ownedBy'] === undefined && (value as Record<string, any>)['owned_by'] === undefined)) return false;
     if (!('modality' in value) || value['modality'] === undefined) return false;
     if (!('pricing' in value) || value['pricing'] === undefined) return false;
@@ -192,20 +248,28 @@ export function ModelFromJSONTyped(json: any, ignoreDiscriminator: boolean): Mod
         
         'id': json['id'],
         'object': json['object'],
-        'created': json['created'] == null ? undefined : json['created'],
+        'created': json['created'],
         'ownedBy': json['owned_by'],
         'author': json['author'] == null ? undefined : json['author'],
         'modality': json['modality'],
         'category': json['category'] == null ? undefined : json['category'],
+        'tier': json['tier'] == null ? undefined : json['tier'],
         'contextLength': json['context_length'] == null ? undefined : json['context_length'],
+        'maxCompletionTokens': json['max_completion_tokens'] == null ? undefined : json['max_completion_tokens'],
         'supportedParameters': json['supported_parameters'] == null ? undefined : json['supported_parameters'],
+        'reasoningEfforts': json['reasoning_efforts'] == null ? undefined : json['reasoning_efforts'],
         'parameters': json['parameters'] == null ? undefined : ((json['parameters'] as Array<any>).map(ModelParametersInnerFromJSON)),
         'pricing': ModelPricingFromJSON(json['pricing']),
         'aliases': json['aliases'] == null ? undefined : json['aliases'],
         'fineTunable': json['fine_tunable'] == null ? undefined : json['fine_tunable'],
         'fineTuneBaseModel': json['fine_tune_base_model'] == null ? undefined : json['fine_tune_base_model'],
+        'hfRepo': json['hf_repo'] == null ? undefined : json['hf_repo'],
+        'usageFactorMilli': json['usage_factor_milli'] == null ? undefined : json['usage_factor_milli'],
         'planCovered': json['plan_covered'] == null ? undefined : json['plan_covered'],
+        'recommendedFor': json['recommended_for'] == null ? undefined : json['recommended_for'],
+        'codingAgentRank': json['coding_agent_rank'] == null ? undefined : json['coding_agent_rank'],
         'status': json['status'],
+        'metrics': json['metrics'] == null ? undefined : ModelMetricsFromJSON(json['metrics']),
     };
 }
 
@@ -227,15 +291,23 @@ export function ModelToJSONTyped(value?: Model | null, ignoreDiscriminator: bool
         'author': value['author'],
         'modality': value['modality'],
         'category': value['category'],
+        'tier': value['tier'],
         'context_length': value['contextLength'],
+        'max_completion_tokens': value['maxCompletionTokens'],
         'supported_parameters': value['supportedParameters'],
+        'reasoning_efforts': value['reasoningEfforts'],
         'parameters': value['parameters'] == null ? undefined : ((value['parameters'] as Array<any>).map(ModelParametersInnerToJSON)),
         'pricing': ModelPricingToJSON(value['pricing']),
         'aliases': value['aliases'],
         'fine_tunable': value['fineTunable'],
         'fine_tune_base_model': value['fineTuneBaseModel'],
+        'hf_repo': value['hfRepo'],
+        'usage_factor_milli': value['usageFactorMilli'],
         'plan_covered': value['planCovered'],
+        'recommended_for': value['recommendedFor'],
+        'coding_agent_rank': value['codingAgentRank'],
         'status': value['status'],
+        'metrics': ModelMetricsToJSON(value['metrics']),
     };
 }
 

package/src/models/ModelList.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/ModelMetrics.ts

@@ -0,0 +1,165 @@
+/* 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';
+/**
+ * Measured performance over a rolling window, from deeprelay's own customer traffic only. A figure is omitted (never zero) when it has too few samples behind it; the counts and minimums are always present so a reader can tell why.
+ * @export
+ * @interface ModelMetrics
+ */
+export interface ModelMetrics {
+    /**
+     * 
+     * @type {number}
+     * @memberof ModelMetrics
+     */
+    windowDays: number;
+    /**
+     * The hour the figures were computed for.
+     * @type {Date}
+     * @memberof ModelMetrics
+     */
+    asOf: Date;
+    /**
+     * 
+     * @type {number}
+     * @memberof ModelMetrics
+     */
+    sampleCount: number;
+    /**
+     * 
+     * @type {number}
+     * @memberof ModelMetrics
+     */
+    streamSampleCount: number;
+    /**
+     * Samples needed before `success_rate` is published.
+     * @type {number}
+     * @memberof ModelMetrics
+     */
+    minSamples: number;
+    /**
+     * Streamed samples needed before the TTFT and tokens/s figures are published.
+     * @type {number}
+     * @memberof ModelMetrics
+     */
+    minStreamSamples: number;
+    /**
+     * Median time to first token, in milliseconds.
+     * @type {number}
+     * @memberof ModelMetrics
+     */
+    ttftP50Ms?: number;
+    /**
+     * Median decode rate after the first token, in tokens per second.
+     * @type {number}
+     * @memberof ModelMetrics
+     */
+    tokensPerSecondP50?: number;
+    /**
+     * 
+     * @type {number}
+     * @memberof ModelMetrics
+     */
+    successRate?: number;
+    /**
+     * True when `sample_count` is below `min_samples`.
+     * @type {boolean}
+     * @memberof ModelMetrics
+     */
+    insufficientData: boolean;
+    /**
+     * What deeprelay can state about the numeric precision this model is served at: `full` (always full precision), `mixed` (some serving capacity is quantized) or `unverified` (not verified; the default).
+     * @type {ModelMetricsPrecisionEnum}
+     * @memberof ModelMetrics
+     */
+    precision: ModelMetricsPrecisionEnum;
+}
+
+
+/**
+ * @export
+ */
+export const ModelMetricsPrecisionEnum = {
+    Full: 'full',
+    Mixed: 'mixed',
+    Unverified: 'unverified'
+} as const;
+export type ModelMetricsPrecisionEnum = typeof ModelMetricsPrecisionEnum[keyof typeof ModelMetricsPrecisionEnum];
+
+
+/**
+ * Check if a given object implements the ModelMetrics interface.
+ */
+export function instanceOfModelMetrics(value: object): value is ModelMetrics {
+    if ((!('windowDays' in (value as Record<string, any>)) && !('window_days' in (value as Record<string, any>))) || ((value as Record<string, any>)['windowDays'] === undefined && (value as Record<string, any>)['window_days'] === undefined)) return false;
+    if ((!('asOf' in (value as Record<string, any>)) && !('as_of' in (value as Record<string, any>))) || ((value as Record<string, any>)['asOf'] === undefined && (value as Record<string, any>)['as_of'] === undefined)) return false;
+    if ((!('sampleCount' in (value as Record<string, any>)) && !('sample_count' in (value as Record<string, any>))) || ((value as Record<string, any>)['sampleCount'] === undefined && (value as Record<string, any>)['sample_count'] === undefined)) return false;
+    if ((!('streamSampleCount' in (value as Record<string, any>)) && !('stream_sample_count' in (value as Record<string, any>))) || ((value as Record<string, any>)['streamSampleCount'] === undefined && (value as Record<string, any>)['stream_sample_count'] === undefined)) return false;
+    if ((!('minSamples' in (value as Record<string, any>)) && !('min_samples' in (value as Record<string, any>))) || ((value as Record<string, any>)['minSamples'] === undefined && (value as Record<string, any>)['min_samples'] === undefined)) return false;
+    if ((!('minStreamSamples' in (value as Record<string, any>)) && !('min_stream_samples' in (value as Record<string, any>))) || ((value as Record<string, any>)['minStreamSamples'] === undefined && (value as Record<string, any>)['min_stream_samples'] === undefined)) return false;
+    if ((!('insufficientData' in (value as Record<string, any>)) && !('insufficient_data' in (value as Record<string, any>))) || ((value as Record<string, any>)['insufficientData'] === undefined && (value as Record<string, any>)['insufficient_data'] === undefined)) return false;
+    if (!('precision' in value) || value['precision'] === undefined) return false;
+    return true;
+}
+
+export function ModelMetricsFromJSON(json: any): ModelMetrics {
+    return ModelMetricsFromJSONTyped(json, false);
+}
+
+export function ModelMetricsFromJSONTyped(json: any, ignoreDiscriminator: boolean): ModelMetrics {
+    if (json == null) {
+        return json;
+    }
+    return {
+        
+        'windowDays': json['window_days'],
+        'asOf': (new Date(json['as_of'])),
+        'sampleCount': json['sample_count'],
+        'streamSampleCount': json['stream_sample_count'],
+        'minSamples': json['min_samples'],
+        'minStreamSamples': json['min_stream_samples'],
+        'ttftP50Ms': json['ttft_p50_ms'] == null ? undefined : json['ttft_p50_ms'],
+        'tokensPerSecondP50': json['tokens_per_second_p50'] == null ? undefined : json['tokens_per_second_p50'],
+        'successRate': json['success_rate'] == null ? undefined : json['success_rate'],
+        'insufficientData': json['insufficient_data'],
+        'precision': json['precision'],
+    };
+}
+
+export function ModelMetricsToJSON(json: any): ModelMetrics {
+    return ModelMetricsToJSONTyped(json, false);
+}
+
+export function ModelMetricsToJSONTyped(value?: ModelMetrics | null, ignoreDiscriminator: boolean = false): any {
+    if (value == null) {
+        return value;
+    }
+
+    return {
+        
+        'window_days': value['windowDays'],
+        'as_of': value['asOf'].toISOString(),
+        'sample_count': value['sampleCount'],
+        'stream_sample_count': value['streamSampleCount'],
+        'min_samples': value['minSamples'],
+        'min_stream_samples': value['minStreamSamples'],
+        'ttft_p50_ms': value['ttftP50Ms'],
+        'tokens_per_second_p50': value['tokensPerSecondP50'],
+        'success_rate': value['successRate'],
+        'insufficient_data': value['insufficientData'],
+        'precision': value['precision'],
+    };
+}
+

package/src/models/ModelParametersInner.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/ModelPricing.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/ModelRate.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/OpenAIErrorEnvelope.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