runbios-sdk 0.2.1-dev.65 → 0.2.1-dev.66

package/dist/index.d.ts

@@ -34,7 +34,7 @@ import { Inference } from './resources/inference.js';
  * SDK version. Sent as part of the User-Agent header.
  * Must match package.json "version" -- enforced by a contract test.
  */
-export declare const VERSION = "0.2.1-dev.65";
+export declare const VERSION = "0.2.1-dev.66";
 export declare class RunBiOS {
     /** Search models, fetch configs, check adapter compatibility. */
     readonly models: Models;
@@ -65,4 +65,4 @@ export { Training } from './resources/training.js';
 export { Wallet } from './resources/wallet.js';
 export { GPU, type GPURecommendation } from './resources/gpu.js';
 export { Inference, validateChatRequest, parseSSE, buildInferenceRequest, contextSizingBasis, type ContextSizingBasis, CONTEXT_DEFAULT_CEILING, CONTEXT_EDITABLE_FLOOR, type ChatMessage, type FunctionTool, type ChatCompletionParams, type ChatCompletionResponse, type ChatCompletionChunk, } from './resources/inference.js';
-export type { BiOSConfig, PaginatedResponse, ApiErrorBody, AvailableGpuAlternative, CapacityMinimumRequirement, InferenceBookingAccepted, Model, ModelSearchParams, ModelSearchResponse, ModelDetailResponse, ModelConfig, Dataset, DatasetListParams, DatasetListResponse, DatasetUploadParams, DatasetPreview, DatasetPreviewParams, DatasetImportHFParams, DatasetRegisterHFParams, DatasetHubSearchParams, DatasetHubPreviewParams, DatasetValidation, DatasetFormatVariant, DatasetFormatSpec, DatasetFormatSpecs, DatasetStorageUsage, TrainingMethod, RLHFAlgorithm, AdapterType, TrainingJobStatus, TrainingStatusFilter, TrainingCreateParams, TrainingListParams, TrainingListResponse, TrainingJob, TrainingMetrics, MetricPoint, MetricGraphConfig, TrainingCheckpoint, TrainingLogs, TrainingLogEntry, TrainingStopResponse, TrainingResumeResponse, CanonicalTrainingRequest, TrainingPreflightDataset, TrainingPreflightWarning, TrainingPreflightResponse, TrainingCapabilityChoice, TrainingConfigFieldCapability, TrainingCapabilities, GPUChoice, WalletBalance, Transaction, TransactionListResponse, TransactionListParams, GPUInfo, GPUPricingResponse, GPUOptionsParams, GPUOption, GPUOptionSuggestion, GPUOptionsResponse, InferenceStatus, InferenceCreateParams, InferenceUpdateParams, InferenceDeployment, InferenceDeploymentSummary, InferenceCreateResponse, InferenceListResponse, InferenceUpdateResponse, InferenceLifecycleResponse, InferenceDeleteResponse, InferenceGPUOptionMarket, InferenceGPUOption, InferenceGPUAlternative, InferenceGPUOptionsParams, InferenceGPUOptionsResponse, InferenceCanonicalRequest, InferencePreflightWarning, InferencePreflightResponse, AdapterCompatibility, AdapterCompatibilityResponse, AdapterCompatibilityParams, SupportedArchitecture, ArchitectureScope, SupportedArchitecturesResponse, ApiKey, ApiKeyScope, ApiKeyIntrospection, Organization, OrgMember, OrgInvite, Workspace, Integration, IntegrationCreateParams, StorageUsage, StorageObject, } from './types.js';
+export type { BiOSConfig, PaginatedResponse, ApiErrorBody, AvailableGpuAlternative, CapacityMinimumRequirement, InferenceBookingAccepted, Model, ModelSearchParams, ModelSearchResponse, ModelDetailResponse, ModelConfig, Dataset, DatasetListParams, DatasetListResponse, DatasetUploadParams, DatasetPreview, DatasetPreviewParams, DatasetImportHFParams, DatasetRegisterHFParams, DatasetHubSearchParams, DatasetHubPreviewParams, DatasetValidation, DatasetFormatVariant, DatasetFormatSpec, DatasetFormatSpecs, DatasetStorageUsage, TrainingMethod, RLHFAlgorithm, AdapterType, TrainingJobStatus, TrainingStatusFilter, TrainingCreateParams, TrainingListParams, TrainingListResponse, TrainingJob, TrainingMetrics, MetricPoint, MetricGraphConfig, TrainingCheckpoint, TrainingLogs, TrainingLogEntry, TrainingStopResponse, TrainingResumeResponse, CanonicalTrainingRequest, TrainingPreflightDataset, TrainingPreflightWarning, TrainingPreflightResponse, TrainingCapabilityChoice, TrainingConfigFieldCapability, TrainingCapabilities, GPUChoice, WalletBalance, Transaction, TransactionListResponse, TransactionListParams, GPUInfo, GPUPricingResponse, GPUOptionsParams, GPUOption, GPUOptionSuggestion, GPUOptionsResponse, InferenceStatus, InferenceCreateParams, InferenceUpdateParams, InferenceDeployment, InferenceDeploymentSummary, InferenceCreateResponse, InferenceListResponse, InferenceUpdateResponse, InferenceLifecycleResponse, InferenceDeleteResponse, InferenceGPUOptionMarket, InferenceGPUOption, InferenceGPUAlternative, InferenceGPUOptionsParams, InferenceGPUOptionsResponse, InferenceCanonicalRequest, InferencePreflightWarning, InferencePreflightResponse, AdapterCompatibility, AdapterCompatibilityResponse, AdapterCompatibilityParams, SupportedArchitecture, ArchitectureScope, SupportedArchitecturesResponse, ApiKey, ApiKeyScope, ApiKeyIntrospection, Organization, OrgMember, OrgInvite, Workspace, Integration, IntegrationCreateParams, StorageUsage, StorageObject, PromptTokensDetails, ChatCompletionUsage, } from './types.js';

package/dist/index.js

@@ -34,7 +34,7 @@ import { Inference } from './resources/inference.js';
  * SDK version. Sent as part of the User-Agent header.
  * Must match package.json "version" -- enforced by a contract test.
  */
-export const VERSION = '0.2.1-dev.65';
+export const VERSION = '0.2.1-dev.66';
 export class RunBiOS {
     /** Search models, fetch configs, check adapter compatibility. */
     models;

package/dist/types.d.ts

@@ -1917,3 +1917,31 @@ export interface StorageObject {
     size_bytes: number;
     created_at: string;
 }
+/**
+ * Prompt-cache slices on a chat completion `usage` object. All three counts
+ * are slices of `prompt_tokens` (cached + writes <= prompt), never additions
+ * to it, and they are the BILLED counts — what the invoice charged, in the
+ * model's public token vocabulary.
+ *
+ * The whole block is absent when the request had no cache activity, so treat
+ * a missing block as "no caching happened", not as zero-with-certainty.
+ * `cached_tokens` is always present inside the block (0 means "wrote but read
+ * nothing back"); the two write counts appear only when non-zero and only on
+ * models that price cache writes separately.
+ */
+export interface PromptTokensDetails {
+    cached_tokens: number;
+    cache_write_5m_tokens?: number;
+    cache_write_1h_tokens?: number;
+}
+/**
+ * Token usage on a `/v1/chat/completions` response (non-stream body, or the
+ * terminal stream chunk when `stream_options.include_usage` was requested).
+ * `total_tokens` is always `prompt_tokens + completion_tokens`.
+ */
+export interface ChatCompletionUsage {
+    prompt_tokens: number;
+    completion_tokens: number;
+    total_tokens: number;
+    prompt_tokens_details?: PromptTokensDetails;
+}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "runbios-sdk",
-  "version": "0.2.1-dev.65",
+  "version": "0.2.1-dev.66",
   "description": "Official TypeScript SDK for the Run BiOS training and deployment platform API",
   "type": "module",
   "main": "./dist/index.js",