weave-typescript 0.8.0 → 0.10.0

package/dist/weaveapi/llmx/v1/pricing.pb.d.ts

@@ -0,0 +1,340 @@
+import { BinaryReader, BinaryWriter } from "@bufbuild/protobuf/wire";
+export declare const protobufPackage = "weaveapi.llmx.v1";
+/** PricingCategory defines the type of AI capability being priced. */
+export declare enum PricingCategory {
+    /** PRICING_CATEGORY_UNSPECIFIED - Default/unknown category */
+    PRICING_CATEGORY_UNSPECIFIED = 0,
+    /**
+     * PRICING_CATEGORY_TEXT - Text generation models (GPT-4, Claude, etc.)
+     * Example: GPT-4o, GPT-4o-mini for chat/completion
+     */
+    PRICING_CATEGORY_TEXT = 1,
+    /**
+     * PRICING_CATEGORY_IMAGE - Image understanding/analysis models
+     * Example: GPT-4-vision analyzing uploaded images
+     */
+    PRICING_CATEGORY_IMAGE = 2,
+    /**
+     * PRICING_CATEGORY_AUDIO - Audio models for speech-to-text and text-to-speech
+     * Example: Whisper for transcription, TTS models for voice synthesis
+     */
+    PRICING_CATEGORY_AUDIO = 3,
+    /**
+     * PRICING_CATEGORY_EMBEDDING - Embedding/vector models
+     * Example: text-embedding-3-large for semantic search
+     */
+    PRICING_CATEGORY_EMBEDDING = 4,
+    /**
+     * PRICING_CATEGORY_FINE_TUNING - Fine-tuning costs for training custom models
+     * Example: Training GPT-4o-mini-2024-07-18 on custom data
+     */
+    PRICING_CATEGORY_FINE_TUNING = 5,
+    /**
+     * PRICING_CATEGORY_TOOL - Built-in or add-on tools and capabilities
+     * Example: Web Search ($2/1k calls), Code Interpreter ($0.03/container)
+     */
+    PRICING_CATEGORY_TOOL = 6,
+    /**
+     * PRICING_CATEGORY_REALTIME - Realtime/streaming API pricing
+     * Example: WebSocket-based continuous interaction models
+     */
+    PRICING_CATEGORY_REALTIME = 7,
+    /**
+     * PRICING_CATEGORY_MODERATION - Content moderation models
+     * Example: OpenAI moderation endpoint for content filtering
+     */
+    PRICING_CATEGORY_MODERATION = 8,
+    /**
+     * PRICING_CATEGORY_LEGACY - Deprecated/legacy models with special pricing
+     * Example: GPT-3.5-turbo-0301 (older version)
+     */
+    PRICING_CATEGORY_LEGACY = 9,
+    UNRECOGNIZED = -1
+}
+export declare function pricingCategoryFromJSON(object: any): PricingCategory;
+export declare function pricingCategoryToJSON(object: PricingCategory): string;
+/** PricingTier represents different service levels with varying costs and SLAs. */
+export declare enum PricingTier {
+    /** PRICING_TIER_UNSPECIFIED - Default/unknown tier */
+    PRICING_TIER_UNSPECIFIED = 0,
+    /**
+     * PRICING_TIER_FREE - Free tier with usage limits
+     * Example: Free API credits for new users
+     */
+    PRICING_TIER_FREE = 1,
+    /**
+     * PRICING_TIER_BATCH - Batch processing with 50% discount, 24-hour turnaround
+     * Example: GPT-4o batch API at $1.25/1M input tokens (vs $2.50 standard)
+     */
+    PRICING_TIER_BATCH = 2,
+    /**
+     * PRICING_TIER_FLEX - Flexible/on-demand tier (between batch and standard)
+     * Example: Variable pricing based on system load
+     */
+    PRICING_TIER_FLEX = 3,
+    /**
+     * PRICING_TIER_STANDARD - Standard/regular pricing tier
+     * Example: GPT-4o at $2.50/1M input tokens for real-time API
+     */
+    PRICING_TIER_STANDARD = 4,
+    /**
+     * PRICING_TIER_PRIORITY - Priority access with higher costs but guaranteed availability
+     * Example: Reserved capacity with SLA guarantees
+     */
+    PRICING_TIER_PRIORITY = 5,
+    /**
+     * PRICING_TIER_ENTERPRISE - Enterprise/custom pricing with negotiated rates
+     * Example: Volume discounts for large organizations
+     */
+    PRICING_TIER_ENTERPRISE = 6,
+    UNRECOGNIZED = -1
+}
+export declare function pricingTierFromJSON(object: any): PricingTier;
+export declare function pricingTierToJSON(object: PricingTier): string;
+/** Operation defines the specific type of usage being priced. */
+export declare enum Operation {
+    /** OPERATION_UNSPECIFIED - Default/unknown operation */
+    OPERATION_UNSPECIFIED = 0,
+    /**
+     * OPERATION_INPUT - Input/prompt tokens sent to the model
+     * Example: User message in chat, $2.50/1M tokens for GPT-4o
+     */
+    OPERATION_INPUT = 1,
+    /**
+     * OPERATION_OUTPUT - Output/completion tokens generated by the model
+     * Example: Model response, $10.00/1M tokens for GPT-4o
+     */
+    OPERATION_OUTPUT = 2,
+    /**
+     * OPERATION_CACHED_INPUT - Cached input tokens (discounted rate for repeated context)
+     * Example: $1.25/1M tokens for GPT-4o cached prompt (50% discount)
+     */
+    OPERATION_CACHED_INPUT = 3,
+    /**
+     * OPERATION_CACHE_READ - Reading from cache (if priced separately from cached input)
+     * Example: Anthropic Claude cache read operations
+     */
+    OPERATION_CACHE_READ = 4,
+    /**
+     * OPERATION_CACHE_WRITE - Writing to cache (creating cached context)
+     * Example: Anthropic Claude cache creation fee
+     */
+    OPERATION_CACHE_WRITE = 5,
+    /**
+     * OPERATION_TRAINING - Fine-tuning training cost (per token or per hour)
+     * Example: $25.00/1M tokens or $3.00/hour for GPT-4o-mini training
+     */
+    OPERATION_TRAINING = 6,
+    /**
+     * OPERATION_STORAGE - General storage costs
+     * Example: Vector database storage for embeddings
+     */
+    OPERATION_STORAGE = 7,
+    /**
+     * OPERATION_CONTEXT_CACHE_STORAGE - Context cache storage charged per time unit
+     * Example: $0.10/GB/hour for maintaining cached contexts
+     */
+    OPERATION_CONTEXT_CACHE_STORAGE = 8,
+    /**
+     * OPERATION_PER_CALL - Fixed price per API call regardless of tokens
+     * Example: $0.00255/call for file search tool
+     */
+    OPERATION_PER_CALL = 9,
+    /**
+     * OPERATION_PER_CONTAINER - Per container/session pricing for isolated environments
+     * Example: $0.03/container for Code Interpreter execution
+     */
+    OPERATION_PER_CONTAINER = 10,
+    UNRECOGNIZED = -1
+}
+export declare function operationFromJSON(object: any): Operation;
+export declare function operationToJSON(object: Operation): string;
+/** PriceUnit specifies how the price is measured/charged. */
+export declare enum PriceUnit {
+    /** PRICE_UNIT_UNSPECIFIED - Default/unknown unit */
+    PRICE_UNIT_UNSPECIFIED = 0,
+    /**
+     * PRICE_UNIT_PER_1M_TOKENS - Price per million tokens (most common for LLMs)
+     * Example: $2.50 per 1M input tokens
+     */
+    PRICE_UNIT_PER_1M_TOKENS = 1,
+    /**
+     * PRICE_UNIT_PER_IMAGE - Price per generated image
+     * Example: DALL-E 3 at $0.040 per 1024x1024 image
+     */
+    PRICE_UNIT_PER_IMAGE = 2,
+    /**
+     * PRICE_UNIT_PER_MINUTE - Price per minute of processing
+     * Example: Whisper at $0.006 per minute of audio
+     */
+    PRICE_UNIT_PER_MINUTE = 3,
+    /**
+     * PRICE_UNIT_PER_1M_CHARACTERS - Price per million characters (for text not tokenized)
+     * Example: Some translation APIs charge per character
+     */
+    PRICE_UNIT_PER_1M_CHARACTERS = 4,
+    /**
+     * PRICE_UNIT_PER_CALL - Fixed price per API call
+     * Example: Moderation at $0.001 per call
+     */
+    PRICE_UNIT_PER_CALL = 5,
+    /**
+     * PRICE_UNIT_PER_GB_DAY - Storage price per GB per day
+     * Example: File search storage at $0.20/GB/day
+     */
+    PRICE_UNIT_PER_GB_DAY = 6,
+    /**
+     * PRICE_UNIT_PER_HOUR - Price per hour (training or runtime)
+     * Example: Fine-tuning at $3.00/hour for small models
+     */
+    PRICE_UNIT_PER_HOUR = 7,
+    /**
+     * PRICE_UNIT_PER_1K_CALLS - Price per thousand API calls
+     * Example: Web search at $2.00 per 1k searches
+     */
+    PRICE_UNIT_PER_1K_CALLS = 8,
+    UNRECOGNIZED = -1
+}
+export declare function priceUnitFromJSON(object: any): PriceUnit;
+export declare function priceUnitToJSON(object: PriceUnit): string;
+/**
+ * Pricing represents the complete pricing structure for a model or service.
+ * A single model may have multiple pricing items for different operations,
+ * tiers, and use cases.
+ */
+export interface Pricing {
+    /**
+     * Currency code for all prices in this message.
+     * Examples: "USD", "EUR", "GBP"
+     */
+    currency: string;
+    /**
+     * List of all pricing entries for this model.
+     * Each item represents a specific price point for a particular
+     * operation/tier combination.
+     * Example: One item for standard input tokens, another for batch input tokens
+     */
+    items: PricingItem[];
+    /**
+     * Additional metadata as key-value pairs.
+     * Common keys:
+     * - "footnote": Additional pricing notes (e.g., "[1] Limited availability")
+     * - "estimated_cost": Estimated costs for transcription models
+     * - "source_url": URL where pricing was sourced from
+     * - "last_updated": ISO 8601 timestamp of last update
+     */
+    metadata: {
+        [key: string]: string;
+    };
+}
+export interface Pricing_MetadataEntry {
+    key: string;
+    value: string;
+}
+/**
+ * TokenBand represents tiered pricing based on token count thresholds.
+ * Some models have different rates based on cumulative token usage.
+ */
+export interface TokenBand {
+    /**
+     * Minimum number of prompt tokens for this pricing tier (inclusive).
+     * Example: 0 for the first tier, 200001 for the "over 200k" tier
+     * Set to 0 if not applicable or for the lowest tier
+     */
+    minPromptTokens: number;
+    /**
+     * Maximum number of prompt tokens for this pricing tier (inclusive).
+     * Example: 200000 for "up to 200k tokens" tier
+     * Set to 0 if unlimited or not applicable
+     */
+    maxPromptTokens: number;
+}
+/**
+ * PricingItem represents a specific price point for a model/service.
+ * Each item captures one pricing scenario with all relevant qualifiers.
+ */
+export interface PricingItem {
+    /**
+     * The type of AI capability this price applies to.
+     * Example: PRICING_CATEGORY_TEXT for GPT-4o chat completion
+     */
+    category: PricingCategory;
+    /**
+     * Service tier affecting price and SLA.
+     * Example: PRICING_TIER_BATCH for 50% discounted async processing
+     */
+    tier: PricingTier;
+    /**
+     * The specific operation being charged for.
+     * Example: OPERATION_INPUT for prompt tokens, OPERATION_OUTPUT for completions
+     */
+    operation: Operation;
+    /**
+     * The numeric cost amount.
+     * Example: 2.50 for "$2.50 per 1M tokens"
+     */
+    price: number;
+    /**
+     * The unit that the price applies to.
+     * Example: PRICE_UNIT_PER_1M_TOKENS for token-based pricing
+     */
+    unit: PriceUnit;
+    /**
+     * Quality level for image generation.
+     * Examples: "standard", "hd" for DALL-E, "low", "medium", "high" for others
+     */
+    quality: string;
+    /**
+     * Image resolution for image generation pricing.
+     * Examples: "1024x1024", "1792x1024", "1024x1792"
+     */
+    resolution: string;
+    /**
+     * Token count thresholds for tiered pricing.
+     * Example: Different rates for prompts over 200k tokens
+     */
+    tokenBand: TokenBand | undefined;
+    /**
+     * Whether output pricing includes hidden reasoning tokens.
+     * Example: true for o1-preview which charges for internal thinking
+     */
+    includesThinkingTokens: boolean;
+    /**
+     * Name of the tool for PRICING_CATEGORY_TOOL items.
+     * Examples: "Web Search", "File Search", "Code Interpreter"
+     */
+    toolName: string;
+    /**
+     * Token type for multi-modal pricing.
+     * Examples: "text" vs "audio" for Whisper transcription models
+     */
+    tokenType: string;
+    /**
+     * Additional human-readable notes about this pricing.
+     * Example: "Discounted pricing with data sharing enabled"
+     */
+    notes: string;
+}
+export declare const Pricing: MessageFns<Pricing>;
+export declare const Pricing_MetadataEntry: MessageFns<Pricing_MetadataEntry>;
+export declare const TokenBand: MessageFns<TokenBand>;
+export declare const PricingItem: MessageFns<PricingItem>;
+type Builtin = Date | Function | Uint8Array | string | number | boolean | undefined;
+export type DeepPartial<T> = T extends Builtin ? T : T extends globalThis.Array<infer U> ? globalThis.Array<DeepPartial<U>> : T extends ReadonlyArray<infer U> ? ReadonlyArray<DeepPartial<U>> : T extends {} ? {
+    [K in keyof T]?: DeepPartial<T[K]>;
+} : Partial<T>;
+type KeysOfUnion<T> = T extends T ? keyof T : never;
+export type Exact<P, I extends P> = P extends Builtin ? P : P & {
+    [K in keyof P]: Exact<P[K], I[K]>;
+} & {
+    [K in Exclude<keyof I, KeysOfUnion<P>>]: never;
+};
+export interface MessageFns<T> {
+    encode(message: T, writer?: BinaryWriter): BinaryWriter;
+    decode(input: BinaryReader | Uint8Array, length?: number): T;
+    fromJSON(object: any): T;
+    toJSON(message: T): unknown;
+    create<I extends Exact<DeepPartial<T>, I>>(base?: I): T;
+    fromPartial<I extends Exact<DeepPartial<T>, I>>(object: I): T;
+}
+export {};