weave-typescript 0.9.0 → 0.10.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/dist/weaveapi/llmx/v1/architecture.pb.d.ts +298 -10
- package/dist/weaveapi/llmx/v1/capabilities.pb.d.ts +219 -1
- package/dist/weaveapi/llmx/v1/capabilities.pb.js +11 -0
- package/dist/weaveapi/llmx/v1/model.pb.d.ts +236 -52
- package/dist/weaveapi/llmx/v1/pricing.pb.d.ts +229 -31
- package/dist/weaveapi/llmx/v1/pricing.pb.js +140 -14
- package/dist/weaveapi/llmx/v1/provider.pb.d.ts +39 -0
- package/dist/weaveapi/llmx/v1/service.pb.d.ts +158 -11
- package/package.json +1 -1
|
@@ -1,82 +1,228 @@
|
|
|
1
1
|
import { BinaryReader, BinaryWriter } from "@bufbuild/protobuf/wire";
|
|
2
2
|
export declare const protobufPackage = "weaveapi.llmx.v1";
|
|
3
|
+
/** PricingCategory defines the type of AI capability being priced. */
|
|
3
4
|
export declare enum PricingCategory {
|
|
5
|
+
/** PRICING_CATEGORY_UNSPECIFIED - Default/unknown category */
|
|
4
6
|
PRICING_CATEGORY_UNSPECIFIED = 0,
|
|
7
|
+
/**
|
|
8
|
+
* PRICING_CATEGORY_TEXT - Text generation models (GPT-4, Claude, etc.)
|
|
9
|
+
* Example: GPT-4o, GPT-4o-mini for chat/completion
|
|
10
|
+
*/
|
|
5
11
|
PRICING_CATEGORY_TEXT = 1,
|
|
12
|
+
/**
|
|
13
|
+
* PRICING_CATEGORY_IMAGE - Image understanding/analysis models
|
|
14
|
+
* Example: GPT-4-vision analyzing uploaded images
|
|
15
|
+
*/
|
|
6
16
|
PRICING_CATEGORY_IMAGE = 2,
|
|
17
|
+
/**
|
|
18
|
+
* PRICING_CATEGORY_AUDIO - Audio models for speech-to-text and text-to-speech
|
|
19
|
+
* Example: Whisper for transcription, TTS models for voice synthesis
|
|
20
|
+
*/
|
|
7
21
|
PRICING_CATEGORY_AUDIO = 3,
|
|
22
|
+
/**
|
|
23
|
+
* PRICING_CATEGORY_EMBEDDING - Embedding/vector models
|
|
24
|
+
* Example: text-embedding-3-large for semantic search
|
|
25
|
+
*/
|
|
8
26
|
PRICING_CATEGORY_EMBEDDING = 4,
|
|
27
|
+
/**
|
|
28
|
+
* PRICING_CATEGORY_FINE_TUNING - Fine-tuning costs for training custom models
|
|
29
|
+
* Example: Training GPT-4o-mini-2024-07-18 on custom data
|
|
30
|
+
*/
|
|
9
31
|
PRICING_CATEGORY_FINE_TUNING = 5,
|
|
10
|
-
/**
|
|
32
|
+
/**
|
|
33
|
+
* PRICING_CATEGORY_TOOL - Built-in or add-on tools and capabilities
|
|
34
|
+
* Example: Web Search ($2/1k calls), Code Interpreter ($0.03/container)
|
|
35
|
+
*/
|
|
11
36
|
PRICING_CATEGORY_TOOL = 6,
|
|
12
|
-
/**
|
|
37
|
+
/**
|
|
38
|
+
* PRICING_CATEGORY_REALTIME - Realtime/streaming API pricing
|
|
39
|
+
* Example: WebSocket-based continuous interaction models
|
|
40
|
+
*/
|
|
13
41
|
PRICING_CATEGORY_REALTIME = 7,
|
|
14
|
-
/**
|
|
42
|
+
/**
|
|
43
|
+
* PRICING_CATEGORY_MODERATION - Content moderation models
|
|
44
|
+
* Example: OpenAI moderation endpoint for content filtering
|
|
45
|
+
*/
|
|
15
46
|
PRICING_CATEGORY_MODERATION = 8,
|
|
16
|
-
/**
|
|
47
|
+
/**
|
|
48
|
+
* PRICING_CATEGORY_LEGACY - Deprecated/legacy models with special pricing
|
|
49
|
+
* Example: GPT-3.5-turbo-0301 (older version)
|
|
50
|
+
*/
|
|
17
51
|
PRICING_CATEGORY_LEGACY = 9,
|
|
18
52
|
UNRECOGNIZED = -1
|
|
19
53
|
}
|
|
20
54
|
export declare function pricingCategoryFromJSON(object: any): PricingCategory;
|
|
21
55
|
export declare function pricingCategoryToJSON(object: PricingCategory): string;
|
|
56
|
+
/** PricingTier represents different service levels with varying costs and SLAs. */
|
|
22
57
|
export declare enum PricingTier {
|
|
58
|
+
/** PRICING_TIER_UNSPECIFIED - Default/unknown tier */
|
|
23
59
|
PRICING_TIER_UNSPECIFIED = 0,
|
|
60
|
+
/**
|
|
61
|
+
* PRICING_TIER_FREE - Free tier with usage limits
|
|
62
|
+
* Example: Free API credits for new users
|
|
63
|
+
*/
|
|
24
64
|
PRICING_TIER_FREE = 1,
|
|
65
|
+
/**
|
|
66
|
+
* PRICING_TIER_BATCH - Batch processing with 50% discount, 24-hour turnaround
|
|
67
|
+
* Example: GPT-4o batch API at $1.25/1M input tokens (vs $2.50 standard)
|
|
68
|
+
*/
|
|
25
69
|
PRICING_TIER_BATCH = 2,
|
|
70
|
+
/**
|
|
71
|
+
* PRICING_TIER_FLEX - Flexible/on-demand tier (between batch and standard)
|
|
72
|
+
* Example: Variable pricing based on system load
|
|
73
|
+
*/
|
|
26
74
|
PRICING_TIER_FLEX = 3,
|
|
75
|
+
/**
|
|
76
|
+
* PRICING_TIER_STANDARD - Standard/regular pricing tier
|
|
77
|
+
* Example: GPT-4o at $2.50/1M input tokens for real-time API
|
|
78
|
+
*/
|
|
27
79
|
PRICING_TIER_STANDARD = 4,
|
|
80
|
+
/**
|
|
81
|
+
* PRICING_TIER_PRIORITY - Priority access with higher costs but guaranteed availability
|
|
82
|
+
* Example: Reserved capacity with SLA guarantees
|
|
83
|
+
*/
|
|
28
84
|
PRICING_TIER_PRIORITY = 5,
|
|
85
|
+
/**
|
|
86
|
+
* PRICING_TIER_ENTERPRISE - Enterprise/custom pricing with negotiated rates
|
|
87
|
+
* Example: Volume discounts for large organizations
|
|
88
|
+
*/
|
|
29
89
|
PRICING_TIER_ENTERPRISE = 6,
|
|
30
90
|
UNRECOGNIZED = -1
|
|
31
91
|
}
|
|
32
92
|
export declare function pricingTierFromJSON(object: any): PricingTier;
|
|
33
93
|
export declare function pricingTierToJSON(object: PricingTier): string;
|
|
94
|
+
/** Operation defines the specific type of usage being priced. */
|
|
34
95
|
export declare enum Operation {
|
|
96
|
+
/** OPERATION_UNSPECIFIED - Default/unknown operation */
|
|
35
97
|
OPERATION_UNSPECIFIED = 0,
|
|
36
|
-
/**
|
|
98
|
+
/**
|
|
99
|
+
* OPERATION_INPUT - Input/prompt tokens sent to the model
|
|
100
|
+
* Example: User message in chat, $2.50/1M tokens for GPT-4o
|
|
101
|
+
*/
|
|
37
102
|
OPERATION_INPUT = 1,
|
|
38
|
-
/**
|
|
103
|
+
/**
|
|
104
|
+
* OPERATION_OUTPUT - Output/completion tokens generated by the model
|
|
105
|
+
* Example: Model response, $10.00/1M tokens for GPT-4o
|
|
106
|
+
*/
|
|
39
107
|
OPERATION_OUTPUT = 2,
|
|
40
|
-
/**
|
|
108
|
+
/**
|
|
109
|
+
* OPERATION_CACHED_INPUT - Cached input tokens (discounted rate for repeated context)
|
|
110
|
+
* Example: $1.25/1M tokens for GPT-4o cached prompt (50% discount)
|
|
111
|
+
*/
|
|
41
112
|
OPERATION_CACHED_INPUT = 3,
|
|
42
|
-
/**
|
|
113
|
+
/**
|
|
114
|
+
* OPERATION_CACHE_READ - Reading from cache (if priced separately from cached input)
|
|
115
|
+
* Example: Anthropic Claude cache read operations
|
|
116
|
+
*/
|
|
43
117
|
OPERATION_CACHE_READ = 4,
|
|
44
|
-
/**
|
|
118
|
+
/**
|
|
119
|
+
* OPERATION_CACHE_WRITE - Writing to cache (creating cached context)
|
|
120
|
+
* Example: Anthropic Claude cache creation fee
|
|
121
|
+
*/
|
|
45
122
|
OPERATION_CACHE_WRITE = 5,
|
|
46
|
-
/**
|
|
123
|
+
/**
|
|
124
|
+
* OPERATION_TRAINING - Fine-tuning training cost (per token or per hour)
|
|
125
|
+
* Example: $25.00/1M tokens or $3.00/hour for GPT-4o-mini training
|
|
126
|
+
*/
|
|
47
127
|
OPERATION_TRAINING = 6,
|
|
48
|
-
/**
|
|
128
|
+
/**
|
|
129
|
+
* OPERATION_STORAGE - General storage costs
|
|
130
|
+
* Example: Vector database storage for embeddings
|
|
131
|
+
*/
|
|
49
132
|
OPERATION_STORAGE = 7,
|
|
50
|
-
/**
|
|
133
|
+
/**
|
|
134
|
+
* OPERATION_CONTEXT_CACHE_STORAGE - Context cache storage charged per time unit
|
|
135
|
+
* Example: $0.10/GB/hour for maintaining cached contexts
|
|
136
|
+
*/
|
|
51
137
|
OPERATION_CONTEXT_CACHE_STORAGE = 8,
|
|
52
|
-
/**
|
|
138
|
+
/**
|
|
139
|
+
* OPERATION_PER_CALL - Fixed price per API call regardless of tokens
|
|
140
|
+
* Example: $0.00255/call for file search tool
|
|
141
|
+
*/
|
|
53
142
|
OPERATION_PER_CALL = 9,
|
|
54
|
-
/**
|
|
143
|
+
/**
|
|
144
|
+
* OPERATION_PER_CONTAINER - Per container/session pricing for isolated environments
|
|
145
|
+
* Example: $0.03/container for Code Interpreter execution
|
|
146
|
+
*/
|
|
55
147
|
OPERATION_PER_CONTAINER = 10,
|
|
56
148
|
UNRECOGNIZED = -1
|
|
57
149
|
}
|
|
58
150
|
export declare function operationFromJSON(object: any): Operation;
|
|
59
151
|
export declare function operationToJSON(object: Operation): string;
|
|
152
|
+
/** PriceUnit specifies how the price is measured/charged. */
|
|
60
153
|
export declare enum PriceUnit {
|
|
154
|
+
/** PRICE_UNIT_UNSPECIFIED - Default/unknown unit */
|
|
61
155
|
PRICE_UNIT_UNSPECIFIED = 0,
|
|
156
|
+
/**
|
|
157
|
+
* PRICE_UNIT_PER_1M_TOKENS - Price per million tokens (most common for LLMs)
|
|
158
|
+
* Example: $2.50 per 1M input tokens
|
|
159
|
+
*/
|
|
62
160
|
PRICE_UNIT_PER_1M_TOKENS = 1,
|
|
161
|
+
/**
|
|
162
|
+
* PRICE_UNIT_PER_IMAGE - Price per generated image
|
|
163
|
+
* Example: DALL-E 3 at $0.040 per 1024x1024 image
|
|
164
|
+
*/
|
|
63
165
|
PRICE_UNIT_PER_IMAGE = 2,
|
|
166
|
+
/**
|
|
167
|
+
* PRICE_UNIT_PER_MINUTE - Price per minute of processing
|
|
168
|
+
* Example: Whisper at $0.006 per minute of audio
|
|
169
|
+
*/
|
|
64
170
|
PRICE_UNIT_PER_MINUTE = 3,
|
|
171
|
+
/**
|
|
172
|
+
* PRICE_UNIT_PER_1M_CHARACTERS - Price per million characters (for text not tokenized)
|
|
173
|
+
* Example: Some translation APIs charge per character
|
|
174
|
+
*/
|
|
65
175
|
PRICE_UNIT_PER_1M_CHARACTERS = 4,
|
|
176
|
+
/**
|
|
177
|
+
* PRICE_UNIT_PER_CALL - Fixed price per API call
|
|
178
|
+
* Example: Moderation at $0.001 per call
|
|
179
|
+
*/
|
|
66
180
|
PRICE_UNIT_PER_CALL = 5,
|
|
181
|
+
/**
|
|
182
|
+
* PRICE_UNIT_PER_GB_DAY - Storage price per GB per day
|
|
183
|
+
* Example: File search storage at $0.20/GB/day
|
|
184
|
+
*/
|
|
67
185
|
PRICE_UNIT_PER_GB_DAY = 6,
|
|
186
|
+
/**
|
|
187
|
+
* PRICE_UNIT_PER_HOUR - Price per hour (training or runtime)
|
|
188
|
+
* Example: Fine-tuning at $3.00/hour for small models
|
|
189
|
+
*/
|
|
68
190
|
PRICE_UNIT_PER_HOUR = 7,
|
|
191
|
+
/**
|
|
192
|
+
* PRICE_UNIT_PER_1K_CALLS - Price per thousand API calls
|
|
193
|
+
* Example: Web search at $2.00 per 1k searches
|
|
194
|
+
*/
|
|
69
195
|
PRICE_UNIT_PER_1K_CALLS = 8,
|
|
70
196
|
UNRECOGNIZED = -1
|
|
71
197
|
}
|
|
72
198
|
export declare function priceUnitFromJSON(object: any): PriceUnit;
|
|
73
199
|
export declare function priceUnitToJSON(object: PriceUnit): string;
|
|
200
|
+
/**
|
|
201
|
+
* Pricing represents the complete pricing structure for a model or service.
|
|
202
|
+
* A single model may have multiple pricing items for different operations,
|
|
203
|
+
* tiers, and use cases.
|
|
204
|
+
*/
|
|
74
205
|
export interface Pricing {
|
|
75
|
-
/**
|
|
206
|
+
/**
|
|
207
|
+
* Currency code for all prices in this message.
|
|
208
|
+
* Examples: "USD", "EUR", "GBP"
|
|
209
|
+
*/
|
|
76
210
|
currency: string;
|
|
77
|
-
/**
|
|
211
|
+
/**
|
|
212
|
+
* List of all pricing entries for this model.
|
|
213
|
+
* Each item represents a specific price point for a particular
|
|
214
|
+
* operation/tier combination.
|
|
215
|
+
* Example: One item for standard input tokens, another for batch input tokens
|
|
216
|
+
*/
|
|
78
217
|
items: PricingItem[];
|
|
79
|
-
/**
|
|
218
|
+
/**
|
|
219
|
+
* Additional metadata as key-value pairs.
|
|
220
|
+
* Common keys:
|
|
221
|
+
* - "footnote": Additional pricing notes (e.g., "[1] Limited availability")
|
|
222
|
+
* - "estimated_cost": Estimated costs for transcription models
|
|
223
|
+
* - "source_url": URL where pricing was sourced from
|
|
224
|
+
* - "last_updated": ISO 8601 timestamp of last update
|
|
225
|
+
*/
|
|
80
226
|
metadata: {
|
|
81
227
|
[key: string]: string;
|
|
82
228
|
};
|
|
@@ -85,36 +231,88 @@ export interface Pricing_MetadataEntry {
|
|
|
85
231
|
key: string;
|
|
86
232
|
value: string;
|
|
87
233
|
}
|
|
234
|
+
/**
|
|
235
|
+
* TokenBand represents tiered pricing based on token count thresholds.
|
|
236
|
+
* Some models have different rates based on cumulative token usage.
|
|
237
|
+
*/
|
|
88
238
|
export interface TokenBand {
|
|
89
|
-
/**
|
|
239
|
+
/**
|
|
240
|
+
* Minimum number of prompt tokens for this pricing tier (inclusive).
|
|
241
|
+
* Example: 0 for the first tier, 200001 for the "over 200k" tier
|
|
242
|
+
* Set to 0 if not applicable or for the lowest tier
|
|
243
|
+
*/
|
|
90
244
|
minPromptTokens: number;
|
|
91
|
-
/**
|
|
245
|
+
/**
|
|
246
|
+
* Maximum number of prompt tokens for this pricing tier (inclusive).
|
|
247
|
+
* Example: 200000 for "up to 200k tokens" tier
|
|
248
|
+
* Set to 0 if unlimited or not applicable
|
|
249
|
+
*/
|
|
92
250
|
maxPromptTokens: number;
|
|
93
251
|
}
|
|
252
|
+
/**
|
|
253
|
+
* PricingItem represents a specific price point for a model/service.
|
|
254
|
+
* Each item captures one pricing scenario with all relevant qualifiers.
|
|
255
|
+
*/
|
|
94
256
|
export interface PricingItem {
|
|
95
|
-
/**
|
|
257
|
+
/**
|
|
258
|
+
* The type of AI capability this price applies to.
|
|
259
|
+
* Example: PRICING_CATEGORY_TEXT for GPT-4o chat completion
|
|
260
|
+
*/
|
|
96
261
|
category: PricingCategory;
|
|
97
|
-
/**
|
|
262
|
+
/**
|
|
263
|
+
* Service tier affecting price and SLA.
|
|
264
|
+
* Example: PRICING_TIER_BATCH for 50% discounted async processing
|
|
265
|
+
*/
|
|
98
266
|
tier: PricingTier;
|
|
99
|
-
/**
|
|
267
|
+
/**
|
|
268
|
+
* The specific operation being charged for.
|
|
269
|
+
* Example: OPERATION_INPUT for prompt tokens, OPERATION_OUTPUT for completions
|
|
270
|
+
*/
|
|
100
271
|
operation: Operation;
|
|
101
|
-
/**
|
|
272
|
+
/**
|
|
273
|
+
* The numeric cost amount.
|
|
274
|
+
* Example: 2.50 for "$2.50 per 1M tokens"
|
|
275
|
+
*/
|
|
102
276
|
price: number;
|
|
103
|
-
/**
|
|
277
|
+
/**
|
|
278
|
+
* The unit that the price applies to.
|
|
279
|
+
* Example: PRICE_UNIT_PER_1M_TOKENS for token-based pricing
|
|
280
|
+
*/
|
|
104
281
|
unit: PriceUnit;
|
|
105
|
-
/**
|
|
282
|
+
/**
|
|
283
|
+
* Quality level for image generation.
|
|
284
|
+
* Examples: "standard", "hd" for DALL-E, "low", "medium", "high" for others
|
|
285
|
+
*/
|
|
106
286
|
quality: string;
|
|
107
|
-
/**
|
|
287
|
+
/**
|
|
288
|
+
* Image resolution for image generation pricing.
|
|
289
|
+
* Examples: "1024x1024", "1792x1024", "1024x1792"
|
|
290
|
+
*/
|
|
108
291
|
resolution: string;
|
|
109
|
-
/**
|
|
292
|
+
/**
|
|
293
|
+
* Token count thresholds for tiered pricing.
|
|
294
|
+
* Example: Different rates for prompts over 200k tokens
|
|
295
|
+
*/
|
|
110
296
|
tokenBand: TokenBand | undefined;
|
|
111
|
-
/**
|
|
297
|
+
/**
|
|
298
|
+
* Whether output pricing includes hidden reasoning tokens.
|
|
299
|
+
* Example: true for o1-preview which charges for internal thinking
|
|
300
|
+
*/
|
|
112
301
|
includesThinkingTokens: boolean;
|
|
113
|
-
/**
|
|
302
|
+
/**
|
|
303
|
+
* Name of the tool for PRICING_CATEGORY_TOOL items.
|
|
304
|
+
* Examples: "Web Search", "File Search", "Code Interpreter"
|
|
305
|
+
*/
|
|
114
306
|
toolName: string;
|
|
115
|
-
/**
|
|
307
|
+
/**
|
|
308
|
+
* Token type for multi-modal pricing.
|
|
309
|
+
* Examples: "text" vs "audio" for Whisper transcription models
|
|
310
|
+
*/
|
|
116
311
|
tokenType: string;
|
|
117
|
-
/**
|
|
312
|
+
/**
|
|
313
|
+
* Additional human-readable notes about this pricing.
|
|
314
|
+
* Example: "Discounted pricing with data sharing enabled"
|
|
315
|
+
*/
|
|
118
316
|
notes: string;
|
|
119
317
|
}
|
|
120
318
|
export declare const Pricing: MessageFns<Pricing>;
|
|
@@ -17,21 +17,55 @@ exports.priceUnitToJSON = priceUnitToJSON;
|
|
|
17
17
|
/* eslint-disable */
|
|
18
18
|
const wire_1 = require("@bufbuild/protobuf/wire");
|
|
19
19
|
exports.protobufPackage = "weaveapi.llmx.v1";
|
|
20
|
+
/** PricingCategory defines the type of AI capability being priced. */
|
|
20
21
|
var PricingCategory;
|
|
21
22
|
(function (PricingCategory) {
|
|
23
|
+
/** PRICING_CATEGORY_UNSPECIFIED - Default/unknown category */
|
|
22
24
|
PricingCategory[PricingCategory["PRICING_CATEGORY_UNSPECIFIED"] = 0] = "PRICING_CATEGORY_UNSPECIFIED";
|
|
25
|
+
/**
|
|
26
|
+
* PRICING_CATEGORY_TEXT - Text generation models (GPT-4, Claude, etc.)
|
|
27
|
+
* Example: GPT-4o, GPT-4o-mini for chat/completion
|
|
28
|
+
*/
|
|
23
29
|
PricingCategory[PricingCategory["PRICING_CATEGORY_TEXT"] = 1] = "PRICING_CATEGORY_TEXT";
|
|
30
|
+
/**
|
|
31
|
+
* PRICING_CATEGORY_IMAGE - Image understanding/analysis models
|
|
32
|
+
* Example: GPT-4-vision analyzing uploaded images
|
|
33
|
+
*/
|
|
24
34
|
PricingCategory[PricingCategory["PRICING_CATEGORY_IMAGE"] = 2] = "PRICING_CATEGORY_IMAGE";
|
|
35
|
+
/**
|
|
36
|
+
* PRICING_CATEGORY_AUDIO - Audio models for speech-to-text and text-to-speech
|
|
37
|
+
* Example: Whisper for transcription, TTS models for voice synthesis
|
|
38
|
+
*/
|
|
25
39
|
PricingCategory[PricingCategory["PRICING_CATEGORY_AUDIO"] = 3] = "PRICING_CATEGORY_AUDIO";
|
|
40
|
+
/**
|
|
41
|
+
* PRICING_CATEGORY_EMBEDDING - Embedding/vector models
|
|
42
|
+
* Example: text-embedding-3-large for semantic search
|
|
43
|
+
*/
|
|
26
44
|
PricingCategory[PricingCategory["PRICING_CATEGORY_EMBEDDING"] = 4] = "PRICING_CATEGORY_EMBEDDING";
|
|
45
|
+
/**
|
|
46
|
+
* PRICING_CATEGORY_FINE_TUNING - Fine-tuning costs for training custom models
|
|
47
|
+
* Example: Training GPT-4o-mini-2024-07-18 on custom data
|
|
48
|
+
*/
|
|
27
49
|
PricingCategory[PricingCategory["PRICING_CATEGORY_FINE_TUNING"] = 5] = "PRICING_CATEGORY_FINE_TUNING";
|
|
28
|
-
/**
|
|
50
|
+
/**
|
|
51
|
+
* PRICING_CATEGORY_TOOL - Built-in or add-on tools and capabilities
|
|
52
|
+
* Example: Web Search ($2/1k calls), Code Interpreter ($0.03/container)
|
|
53
|
+
*/
|
|
29
54
|
PricingCategory[PricingCategory["PRICING_CATEGORY_TOOL"] = 6] = "PRICING_CATEGORY_TOOL";
|
|
30
|
-
/**
|
|
55
|
+
/**
|
|
56
|
+
* PRICING_CATEGORY_REALTIME - Realtime/streaming API pricing
|
|
57
|
+
* Example: WebSocket-based continuous interaction models
|
|
58
|
+
*/
|
|
31
59
|
PricingCategory[PricingCategory["PRICING_CATEGORY_REALTIME"] = 7] = "PRICING_CATEGORY_REALTIME";
|
|
32
|
-
/**
|
|
60
|
+
/**
|
|
61
|
+
* PRICING_CATEGORY_MODERATION - Content moderation models
|
|
62
|
+
* Example: OpenAI moderation endpoint for content filtering
|
|
63
|
+
*/
|
|
33
64
|
PricingCategory[PricingCategory["PRICING_CATEGORY_MODERATION"] = 8] = "PRICING_CATEGORY_MODERATION";
|
|
34
|
-
/**
|
|
65
|
+
/**
|
|
66
|
+
* PRICING_CATEGORY_LEGACY - Deprecated/legacy models with special pricing
|
|
67
|
+
* Example: GPT-3.5-turbo-0301 (older version)
|
|
68
|
+
*/
|
|
35
69
|
PricingCategory[PricingCategory["PRICING_CATEGORY_LEGACY"] = 9] = "PRICING_CATEGORY_LEGACY";
|
|
36
70
|
PricingCategory[PricingCategory["UNRECOGNIZED"] = -1] = "UNRECOGNIZED";
|
|
37
71
|
})(PricingCategory || (exports.PricingCategory = PricingCategory = {}));
|
|
@@ -100,14 +134,40 @@ function pricingCategoryToJSON(object) {
|
|
|
100
134
|
return "UNRECOGNIZED";
|
|
101
135
|
}
|
|
102
136
|
}
|
|
137
|
+
/** PricingTier represents different service levels with varying costs and SLAs. */
|
|
103
138
|
var PricingTier;
|
|
104
139
|
(function (PricingTier) {
|
|
140
|
+
/** PRICING_TIER_UNSPECIFIED - Default/unknown tier */
|
|
105
141
|
PricingTier[PricingTier["PRICING_TIER_UNSPECIFIED"] = 0] = "PRICING_TIER_UNSPECIFIED";
|
|
142
|
+
/**
|
|
143
|
+
* PRICING_TIER_FREE - Free tier with usage limits
|
|
144
|
+
* Example: Free API credits for new users
|
|
145
|
+
*/
|
|
106
146
|
PricingTier[PricingTier["PRICING_TIER_FREE"] = 1] = "PRICING_TIER_FREE";
|
|
147
|
+
/**
|
|
148
|
+
* PRICING_TIER_BATCH - Batch processing with 50% discount, 24-hour turnaround
|
|
149
|
+
* Example: GPT-4o batch API at $1.25/1M input tokens (vs $2.50 standard)
|
|
150
|
+
*/
|
|
107
151
|
PricingTier[PricingTier["PRICING_TIER_BATCH"] = 2] = "PRICING_TIER_BATCH";
|
|
152
|
+
/**
|
|
153
|
+
* PRICING_TIER_FLEX - Flexible/on-demand tier (between batch and standard)
|
|
154
|
+
* Example: Variable pricing based on system load
|
|
155
|
+
*/
|
|
108
156
|
PricingTier[PricingTier["PRICING_TIER_FLEX"] = 3] = "PRICING_TIER_FLEX";
|
|
157
|
+
/**
|
|
158
|
+
* PRICING_TIER_STANDARD - Standard/regular pricing tier
|
|
159
|
+
* Example: GPT-4o at $2.50/1M input tokens for real-time API
|
|
160
|
+
*/
|
|
109
161
|
PricingTier[PricingTier["PRICING_TIER_STANDARD"] = 4] = "PRICING_TIER_STANDARD";
|
|
162
|
+
/**
|
|
163
|
+
* PRICING_TIER_PRIORITY - Priority access with higher costs but guaranteed availability
|
|
164
|
+
* Example: Reserved capacity with SLA guarantees
|
|
165
|
+
*/
|
|
110
166
|
PricingTier[PricingTier["PRICING_TIER_PRIORITY"] = 5] = "PRICING_TIER_PRIORITY";
|
|
167
|
+
/**
|
|
168
|
+
* PRICING_TIER_ENTERPRISE - Enterprise/custom pricing with negotiated rates
|
|
169
|
+
* Example: Volume discounts for large organizations
|
|
170
|
+
*/
|
|
111
171
|
PricingTier[PricingTier["PRICING_TIER_ENTERPRISE"] = 6] = "PRICING_TIER_ENTERPRISE";
|
|
112
172
|
PricingTier[PricingTier["UNRECOGNIZED"] = -1] = "UNRECOGNIZED";
|
|
113
173
|
})(PricingTier || (exports.PricingTier = PricingTier = {}));
|
|
@@ -161,28 +221,60 @@ function pricingTierToJSON(object) {
|
|
|
161
221
|
return "UNRECOGNIZED";
|
|
162
222
|
}
|
|
163
223
|
}
|
|
224
|
+
/** Operation defines the specific type of usage being priced. */
|
|
164
225
|
var Operation;
|
|
165
226
|
(function (Operation) {
|
|
227
|
+
/** OPERATION_UNSPECIFIED - Default/unknown operation */
|
|
166
228
|
Operation[Operation["OPERATION_UNSPECIFIED"] = 0] = "OPERATION_UNSPECIFIED";
|
|
167
|
-
/**
|
|
229
|
+
/**
|
|
230
|
+
* OPERATION_INPUT - Input/prompt tokens sent to the model
|
|
231
|
+
* Example: User message in chat, $2.50/1M tokens for GPT-4o
|
|
232
|
+
*/
|
|
168
233
|
Operation[Operation["OPERATION_INPUT"] = 1] = "OPERATION_INPUT";
|
|
169
|
-
/**
|
|
234
|
+
/**
|
|
235
|
+
* OPERATION_OUTPUT - Output/completion tokens generated by the model
|
|
236
|
+
* Example: Model response, $10.00/1M tokens for GPT-4o
|
|
237
|
+
*/
|
|
170
238
|
Operation[Operation["OPERATION_OUTPUT"] = 2] = "OPERATION_OUTPUT";
|
|
171
|
-
/**
|
|
239
|
+
/**
|
|
240
|
+
* OPERATION_CACHED_INPUT - Cached input tokens (discounted rate for repeated context)
|
|
241
|
+
* Example: $1.25/1M tokens for GPT-4o cached prompt (50% discount)
|
|
242
|
+
*/
|
|
172
243
|
Operation[Operation["OPERATION_CACHED_INPUT"] = 3] = "OPERATION_CACHED_INPUT";
|
|
173
|
-
/**
|
|
244
|
+
/**
|
|
245
|
+
* OPERATION_CACHE_READ - Reading from cache (if priced separately from cached input)
|
|
246
|
+
* Example: Anthropic Claude cache read operations
|
|
247
|
+
*/
|
|
174
248
|
Operation[Operation["OPERATION_CACHE_READ"] = 4] = "OPERATION_CACHE_READ";
|
|
175
|
-
/**
|
|
249
|
+
/**
|
|
250
|
+
* OPERATION_CACHE_WRITE - Writing to cache (creating cached context)
|
|
251
|
+
* Example: Anthropic Claude cache creation fee
|
|
252
|
+
*/
|
|
176
253
|
Operation[Operation["OPERATION_CACHE_WRITE"] = 5] = "OPERATION_CACHE_WRITE";
|
|
177
|
-
/**
|
|
254
|
+
/**
|
|
255
|
+
* OPERATION_TRAINING - Fine-tuning training cost (per token or per hour)
|
|
256
|
+
* Example: $25.00/1M tokens or $3.00/hour for GPT-4o-mini training
|
|
257
|
+
*/
|
|
178
258
|
Operation[Operation["OPERATION_TRAINING"] = 6] = "OPERATION_TRAINING";
|
|
179
|
-
/**
|
|
259
|
+
/**
|
|
260
|
+
* OPERATION_STORAGE - General storage costs
|
|
261
|
+
* Example: Vector database storage for embeddings
|
|
262
|
+
*/
|
|
180
263
|
Operation[Operation["OPERATION_STORAGE"] = 7] = "OPERATION_STORAGE";
|
|
181
|
-
/**
|
|
264
|
+
/**
|
|
265
|
+
* OPERATION_CONTEXT_CACHE_STORAGE - Context cache storage charged per time unit
|
|
266
|
+
* Example: $0.10/GB/hour for maintaining cached contexts
|
|
267
|
+
*/
|
|
182
268
|
Operation[Operation["OPERATION_CONTEXT_CACHE_STORAGE"] = 8] = "OPERATION_CONTEXT_CACHE_STORAGE";
|
|
183
|
-
/**
|
|
269
|
+
/**
|
|
270
|
+
* OPERATION_PER_CALL - Fixed price per API call regardless of tokens
|
|
271
|
+
* Example: $0.00255/call for file search tool
|
|
272
|
+
*/
|
|
184
273
|
Operation[Operation["OPERATION_PER_CALL"] = 9] = "OPERATION_PER_CALL";
|
|
185
|
-
/**
|
|
274
|
+
/**
|
|
275
|
+
* OPERATION_PER_CONTAINER - Per container/session pricing for isolated environments
|
|
276
|
+
* Example: $0.03/container for Code Interpreter execution
|
|
277
|
+
*/
|
|
186
278
|
Operation[Operation["OPERATION_PER_CONTAINER"] = 10] = "OPERATION_PER_CONTAINER";
|
|
187
279
|
Operation[Operation["UNRECOGNIZED"] = -1] = "UNRECOGNIZED";
|
|
188
280
|
})(Operation || (exports.Operation = Operation = {}));
|
|
@@ -256,16 +348,50 @@ function operationToJSON(object) {
|
|
|
256
348
|
return "UNRECOGNIZED";
|
|
257
349
|
}
|
|
258
350
|
}
|
|
351
|
+
/** PriceUnit specifies how the price is measured/charged. */
|
|
259
352
|
var PriceUnit;
|
|
260
353
|
(function (PriceUnit) {
|
|
354
|
+
/** PRICE_UNIT_UNSPECIFIED - Default/unknown unit */
|
|
261
355
|
PriceUnit[PriceUnit["PRICE_UNIT_UNSPECIFIED"] = 0] = "PRICE_UNIT_UNSPECIFIED";
|
|
356
|
+
/**
|
|
357
|
+
* PRICE_UNIT_PER_1M_TOKENS - Price per million tokens (most common for LLMs)
|
|
358
|
+
* Example: $2.50 per 1M input tokens
|
|
359
|
+
*/
|
|
262
360
|
PriceUnit[PriceUnit["PRICE_UNIT_PER_1M_TOKENS"] = 1] = "PRICE_UNIT_PER_1M_TOKENS";
|
|
361
|
+
/**
|
|
362
|
+
* PRICE_UNIT_PER_IMAGE - Price per generated image
|
|
363
|
+
* Example: DALL-E 3 at $0.040 per 1024x1024 image
|
|
364
|
+
*/
|
|
263
365
|
PriceUnit[PriceUnit["PRICE_UNIT_PER_IMAGE"] = 2] = "PRICE_UNIT_PER_IMAGE";
|
|
366
|
+
/**
|
|
367
|
+
* PRICE_UNIT_PER_MINUTE - Price per minute of processing
|
|
368
|
+
* Example: Whisper at $0.006 per minute of audio
|
|
369
|
+
*/
|
|
264
370
|
PriceUnit[PriceUnit["PRICE_UNIT_PER_MINUTE"] = 3] = "PRICE_UNIT_PER_MINUTE";
|
|
371
|
+
/**
|
|
372
|
+
* PRICE_UNIT_PER_1M_CHARACTERS - Price per million characters (for text not tokenized)
|
|
373
|
+
* Example: Some translation APIs charge per character
|
|
374
|
+
*/
|
|
265
375
|
PriceUnit[PriceUnit["PRICE_UNIT_PER_1M_CHARACTERS"] = 4] = "PRICE_UNIT_PER_1M_CHARACTERS";
|
|
376
|
+
/**
|
|
377
|
+
* PRICE_UNIT_PER_CALL - Fixed price per API call
|
|
378
|
+
* Example: Moderation at $0.001 per call
|
|
379
|
+
*/
|
|
266
380
|
PriceUnit[PriceUnit["PRICE_UNIT_PER_CALL"] = 5] = "PRICE_UNIT_PER_CALL";
|
|
381
|
+
/**
|
|
382
|
+
* PRICE_UNIT_PER_GB_DAY - Storage price per GB per day
|
|
383
|
+
* Example: File search storage at $0.20/GB/day
|
|
384
|
+
*/
|
|
267
385
|
PriceUnit[PriceUnit["PRICE_UNIT_PER_GB_DAY"] = 6] = "PRICE_UNIT_PER_GB_DAY";
|
|
386
|
+
/**
|
|
387
|
+
* PRICE_UNIT_PER_HOUR - Price per hour (training or runtime)
|
|
388
|
+
* Example: Fine-tuning at $3.00/hour for small models
|
|
389
|
+
*/
|
|
268
390
|
PriceUnit[PriceUnit["PRICE_UNIT_PER_HOUR"] = 7] = "PRICE_UNIT_PER_HOUR";
|
|
391
|
+
/**
|
|
392
|
+
* PRICE_UNIT_PER_1K_CALLS - Price per thousand API calls
|
|
393
|
+
* Example: Web search at $2.00 per 1k searches
|
|
394
|
+
*/
|
|
269
395
|
PriceUnit[PriceUnit["PRICE_UNIT_PER_1K_CALLS"] = 8] = "PRICE_UNIT_PER_1K_CALLS";
|
|
270
396
|
PriceUnit[PriceUnit["UNRECOGNIZED"] = -1] = "UNRECOGNIZED";
|
|
271
397
|
})(PriceUnit || (exports.PriceUnit = PriceUnit = {}));
|
|
@@ -1,14 +1,53 @@
|
|
|
1
1
|
import { BinaryReader, BinaryWriter } from "@bufbuild/protobuf/wire";
|
|
2
2
|
export declare const protobufPackage = "weaveapi.llmx.v1";
|
|
3
|
+
/** Provider represents an AI model provider/vendor organization. */
|
|
3
4
|
export interface Provider {
|
|
5
|
+
/**
|
|
6
|
+
* Unique internal identifier for this provider.
|
|
7
|
+
* Example: "pvd_01234567-89ab-cdef-0123-456789abcdef"
|
|
8
|
+
*/
|
|
4
9
|
id: string;
|
|
10
|
+
/**
|
|
11
|
+
* URL-friendly identifier for routing.
|
|
12
|
+
* Examples: "openai", "anthropic", "google", "meta"
|
|
13
|
+
* Used in paths like /providers/openai
|
|
14
|
+
*/
|
|
5
15
|
slug: string;
|
|
16
|
+
/**
|
|
17
|
+
* Display name of the provider.
|
|
18
|
+
* Examples: "OpenAI", "Anthropic", "Google AI", "Meta AI"
|
|
19
|
+
*/
|
|
6
20
|
name: string;
|
|
21
|
+
/**
|
|
22
|
+
* Description of the provider and their focus.
|
|
23
|
+
* Example: "OpenAI develops general-purpose AI models including GPT-4,
|
|
24
|
+
* DALL-E, and Whisper, focusing on safe and beneficial AI."
|
|
25
|
+
*/
|
|
7
26
|
description: string;
|
|
27
|
+
/**
|
|
28
|
+
* Main website URL.
|
|
29
|
+
* Example: "https://openai.com"
|
|
30
|
+
*/
|
|
8
31
|
websiteUrl: string;
|
|
32
|
+
/**
|
|
33
|
+
* API documentation URL.
|
|
34
|
+
* Example: "https://platform.openai.com/docs"
|
|
35
|
+
*/
|
|
9
36
|
documentationUrl: string;
|
|
37
|
+
/**
|
|
38
|
+
* URL to provider's logo image.
|
|
39
|
+
* Example: "https://assets.example.com/logos/openai.svg"
|
|
40
|
+
*/
|
|
10
41
|
logoUrl: string;
|
|
42
|
+
/**
|
|
43
|
+
* Whether provider is currently operational.
|
|
44
|
+
* Example: false if provider has shut down or paused service
|
|
45
|
+
*/
|
|
11
46
|
isActive: boolean;
|
|
47
|
+
/**
|
|
48
|
+
* Number of models offered by this provider.
|
|
49
|
+
* Example: 25 for a provider with 25 different models
|
|
50
|
+
*/
|
|
12
51
|
modelCount: number;
|
|
13
52
|
}
|
|
14
53
|
export declare const Provider: MessageFns<Provider>;
|