@mate-academy/llm-gateway 6.2.1 → 7.0.0

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/README.md CHANGED
@@ -621,7 +621,7 @@ const customGPT5Model = {
621
621
  config: {
622
622
  ...gpt5Model.config,
623
623
  temperature: 1, // Note: GPT-5 only supports temperature=1
624
- reasoning_effort: 'high', // 'minimal' | 'medium' | 'high'
624
+ reasoning_effort: 'high', // 'minimal' | 'low' | 'medium' | 'high' | 'xhigh'
625
625
  verbosity: 'low', // 'low' | 'medium' | 'high'
626
626
  }
627
627
  };
@@ -634,26 +634,26 @@ const customGPT51Model = {
634
634
  config: {
635
635
  ...gpt51Model.config,
636
636
  temperature: 1, // Note: GPT-5.1 only supports temperature=1
637
- reasoning_effort: 'none', // 'none' | 'minimal' | 'medium' | 'high'
637
+ reasoning_effort: 'none', // 'none' | 'minimal' | 'low' | 'medium' | 'high' | 'xhigh'
638
638
  verbosity: 'low', // 'low' | 'medium' | 'high'
639
639
  }
640
640
  };
641
641
  ```
642
642
 
643
- **Note:** All GPT-5 models require `temperature: 1` and cannot be changed. GPT-5, GPT-5-MINI, and GPT-5-NANO support reasoning_effort values of 'minimal', 'medium', or 'high'. GPT-5.1 and GPT-5.2 additionally supports 'none' as a reasoning_effort value.
643
+ **Note:** All GPT-5 models require `temperature: 1` and cannot be changed. GPT-5, GPT-5-MINI, and GPT-5-NANO support reasoning_effort values of 'minimal', 'low', 'medium', 'high', or 'xhigh'. GPT-5.1, GPT-5.2, and GPT-5.4 additionally support 'none'.
644
644
 
645
- **Gemini 3 Pro Preview** has special configuration options:
645
+ **Gemini 3.x thinking models** (gemini-3-flash-preview, gemini-3.1-pro-preview) have special configuration options:
646
646
 
647
647
  ```typescript
648
648
  import { ThinkingLevel } from '@google/genai';
649
649
 
650
- const gemini3Model = models[GoogleGenerativeAIModelNames.GEMINI_3_PRO_PREVIEW];
650
+ const gemini31Model = models[GoogleGenerativeAIModelNames.GEMINI_3_1_PRO_PREVIEW];
651
651
 
652
- const customGemini3Model = {
653
- ...gemini3Model,
652
+ const customGemini31Model = {
653
+ ...gemini31Model,
654
654
  config: {
655
- ...gemini3Model.config,
656
- temperature: 1, // Note: Gemini 3 Pro Preview only supports temperature=1
655
+ ...gemini31Model.config,
656
+ temperature: 1, // Note: Gemini 3.x thinking models only support temperature=1
657
657
  thinkingConfig: {
658
658
  thinkingLevel: ThinkingLevel.HIGH, // LOW | HIGH | THINKING_LEVEL_UNSPECIFIED
659
659
  },
@@ -661,7 +661,7 @@ const customGemini3Model = {
661
661
  };
662
662
  ```
663
663
 
664
- **Note:** Gemini 3 Pro Preview requires `temperature: 1` and cannot be changed. The `thinkingConfig` allows you to control the model's reasoning depth.
664
+ **Note:** Gemini 3.x thinking models require `temperature: 1` and cannot be changed. The `thinkingConfig` allows you to control the model's reasoning depth.
665
665
 
666
666
  #### File-based Assistance
667
667
 
@@ -1311,9 +1311,11 @@ Supports all service types: completion, assistance, speech-to-text, and text-to-
1311
1311
  |-------|---------|-----------|------------|-------|
1312
1312
  | gpt-4.1 | Completion, Assistance | 1M+ | 32K | Extended context window |
1313
1313
  | gpt-4.1-mini | Completion, Assistance | 1M+ | 32K | Cost-effective extended context |
1314
+ | gpt-4.1-nano | Completion, Assistance | 1M+ | 32K | Fastest, most cost-efficient GPT-4.1 |
1314
1315
  | gpt-5 | Completion, Assistance | 400K | 128K | Advanced reasoning, requires temperature=1 |
1315
1316
  | gpt-5.1 | Completion, Assistance | 400K | 128K | Advanced reasoning with 'none' reasoning_effort support, requires temperature=1 |
1316
1317
  | gpt-5.2 | Completion, Assistance | 400K | 128K | Advanced reasoning with 'none' reasoning_effort support, requires temperature=1 |
1318
+ | gpt-5.4 | Completion, Assistance | 1M+ | 128K | Flagship model, supports 'none'/'xhigh' reasoning_effort, requires temperature=1 |
1317
1319
  | gpt-5-mini | Completion, Assistance | 400K | 128K | Smaller GPT-5 variant, requires temperature=1 |
1318
1320
  | gpt-5-nano | Completion, Assistance | 400K | 128K | Fastest GPT-5 variant, requires temperature=1 |
1319
1321
  | gpt-4o-transcribe | Speech-to-Text | 16K | 2K | Optimized for transcription |
@@ -1332,13 +1334,16 @@ Supports completion, assistance, speech-to-text, and text-to-speech APIs through
1332
1334
  | gemini-2.5-flash | Completion, Assistance, Speech-to-Text | 1M+ | 65K | Fast, cost-effective, supports caching for long context |
1333
1335
  | gemini-2.5-flash-lite | Completion, Assistance, Speech-to-Text | 1M+ | 65K | Fastest, most cost-effective, supports caching |
1334
1336
  | gemini-2.5-pro | Completion, Assistance, Speech-to-Text | 1M+ | 65K | Advanced reasoning, supports caching for long context |
1335
- | gemini-3-pro-preview | Completion, Assistance, Speech-to-Text | 1M+ | 65K | Advanced reasoning with thinking capabilities, requires temperature=1, supports configurable thinking levels |
1337
+ | gemini-3-flash-preview | Completion, Assistance, Speech-to-Text | 1M+ | 65K | Thinking model, requires temperature=1, configurable thinking levels |
1338
+ | gemini-3.1-flash-lite-preview | Completion, Assistance, Speech-to-Text | 1M+ | 65K | Most cost-efficient Gemini 3.x model |
1339
+ | gemini-3.1-pro-preview | Completion, Assistance, Speech-to-Text | 1M+ | 65K | Advanced reasoning with thinking capabilities, requires temperature=1 |
1336
1340
  | gemini-2.5-flash-preview-tts | Text-to-Speech | 8K | 16K | Preview TTS model with flash performance |
1337
1341
  | gemini-2.5-pro-preview-tts | Text-to-Speech | 8K | 16K | Preview TTS model with pro capabilities |
1338
1342
 
1339
1343
  **Note:** Google Generative AI models support context caching for content longer than 32,768 tokens, which can significantly reduce costs for repeated queries on the same large context.
1340
1344
 
1341
- **Gemini 3 Pro Preview Configuration:**
1345
+ **Gemini 3.x Thinking Model Configuration:**
1346
+ - Applies to: gemini-3-flash-preview, gemini-3.1-pro-preview
1342
1347
  - Requires `temperature: 1` (cannot be changed)
1343
1348
  - Supports `thinkingConfig` with `thinkingLevel` property (LOW, HIGH, THINKING_LEVEL_UNSPECIFIED)
1344
1349
  - Default thinking level is LOW
@@ -50,6 +50,27 @@ export declare const LLM_SERVICE_MODELS: {
50
50
  readonly vision: true;
51
51
  };
52
52
  };
53
+ readonly "gpt-4.1-nano": {
54
+ readonly name: import("./providers").OpenAIModelNames.GPT_4_1_NANO;
55
+ readonly limits: {
56
+ readonly maxInputTokens: 1047576;
57
+ readonly maxOutputTokens: 32768;
58
+ };
59
+ readonly config: {
60
+ readonly temperature: 0.2;
61
+ };
62
+ readonly pricing: {
63
+ readonly getPriceForTextInput: (tokens: number) => number;
64
+ readonly getPriceForTextOutput: (tokens: number) => number;
65
+ readonly getPriceForCachedTextInput: (tokens: number) => number;
66
+ readonly getPriceForAudioInput: () => number;
67
+ readonly getPriceForAudioOutput: () => number;
68
+ readonly currency: "USD";
69
+ };
70
+ readonly capabilities: {
71
+ readonly vision: true;
72
+ };
73
+ };
53
74
  readonly "gpt-5": {
54
75
  readonly name: import("./providers").OpenAIModelNames.GPT_5;
55
76
  readonly limits: {
@@ -119,6 +140,29 @@ export declare const LLM_SERVICE_MODELS: {
119
140
  readonly vision: true;
120
141
  };
121
142
  };
143
+ readonly "gpt-5.4": {
144
+ readonly name: import("./providers").OpenAIModelNames.GPT_5_4;
145
+ readonly limits: {
146
+ readonly maxInputTokens: 1050000;
147
+ readonly maxOutputTokens: 128000;
148
+ };
149
+ readonly config: {
150
+ readonly temperature: 1;
151
+ readonly reasoning_effort: "none";
152
+ readonly verbosity: "medium";
153
+ };
154
+ readonly pricing: {
155
+ readonly getPriceForTextInput: (tokens: number) => number;
156
+ readonly getPriceForTextOutput: (tokens: number) => number;
157
+ readonly getPriceForCachedTextInput: (tokens: number) => number;
158
+ readonly getPriceForAudioInput: () => number;
159
+ readonly getPriceForAudioOutput: () => number;
160
+ readonly currency: "USD";
161
+ };
162
+ readonly capabilities: {
163
+ readonly vision: true;
164
+ };
165
+ };
122
166
  readonly "gpt-5-mini": {
123
167
  readonly name: import("./providers").OpenAIModelNames.GPT_5_MINI;
124
168
  readonly limits: {
@@ -249,7 +293,7 @@ export declare const LLM_SERVICE_MODELS: {
249
293
  readonly vision: false;
250
294
  };
251
295
  };
252
- }, import("./providers").OpenAIModelNames.GPT_4_1 | import("./providers").OpenAIModelNames.GPT_4_1_MINI | import("./providers").OpenAIModelNames.GPT_5 | import("./providers").OpenAIModelNames.GPT_5_1 | import("./providers").OpenAIModelNames.GPT_5_2 | import("./providers").OpenAIModelNames.GPT_5_MINI | import("./providers").OpenAIModelNames.GPT_5_NANO>;
296
+ }, import("./providers").OpenAIModelNames.GPT_4_1 | import("./providers").OpenAIModelNames.GPT_4_1_MINI | import("./providers").OpenAIModelNames.GPT_4_1_NANO | import("./providers").OpenAIModelNames.GPT_5 | import("./providers").OpenAIModelNames.GPT_5_1 | import("./providers").OpenAIModelNames.GPT_5_2 | import("./providers").OpenAIModelNames.GPT_5_4 | import("./providers").OpenAIModelNames.GPT_5_MINI | import("./providers").OpenAIModelNames.GPT_5_NANO>;
253
297
  readonly assistance: Pick<{
254
298
  readonly "gpt-4.1": {
255
299
  readonly name: import("./providers").OpenAIModelNames.GPT_4_1;
@@ -293,6 +337,27 @@ export declare const LLM_SERVICE_MODELS: {
293
337
  readonly vision: true;
294
338
  };
295
339
  };
340
+ readonly "gpt-4.1-nano": {
341
+ readonly name: import("./providers").OpenAIModelNames.GPT_4_1_NANO;
342
+ readonly limits: {
343
+ readonly maxInputTokens: 1047576;
344
+ readonly maxOutputTokens: 32768;
345
+ };
346
+ readonly config: {
347
+ readonly temperature: 0.2;
348
+ };
349
+ readonly pricing: {
350
+ readonly getPriceForTextInput: (tokens: number) => number;
351
+ readonly getPriceForTextOutput: (tokens: number) => number;
352
+ readonly getPriceForCachedTextInput: (tokens: number) => number;
353
+ readonly getPriceForAudioInput: () => number;
354
+ readonly getPriceForAudioOutput: () => number;
355
+ readonly currency: "USD";
356
+ };
357
+ readonly capabilities: {
358
+ readonly vision: true;
359
+ };
360
+ };
296
361
  readonly "gpt-5": {
297
362
  readonly name: import("./providers").OpenAIModelNames.GPT_5;
298
363
  readonly limits: {
@@ -362,6 +427,29 @@ export declare const LLM_SERVICE_MODELS: {
362
427
  readonly vision: true;
363
428
  };
364
429
  };
430
+ readonly "gpt-5.4": {
431
+ readonly name: import("./providers").OpenAIModelNames.GPT_5_4;
432
+ readonly limits: {
433
+ readonly maxInputTokens: 1050000;
434
+ readonly maxOutputTokens: 128000;
435
+ };
436
+ readonly config: {
437
+ readonly temperature: 1;
438
+ readonly reasoning_effort: "none";
439
+ readonly verbosity: "medium";
440
+ };
441
+ readonly pricing: {
442
+ readonly getPriceForTextInput: (tokens: number) => number;
443
+ readonly getPriceForTextOutput: (tokens: number) => number;
444
+ readonly getPriceForCachedTextInput: (tokens: number) => number;
445
+ readonly getPriceForAudioInput: () => number;
446
+ readonly getPriceForAudioOutput: () => number;
447
+ readonly currency: "USD";
448
+ };
449
+ readonly capabilities: {
450
+ readonly vision: true;
451
+ };
452
+ };
365
453
  readonly "gpt-5-mini": {
366
454
  readonly name: import("./providers").OpenAIModelNames.GPT_5_MINI;
367
455
  readonly limits: {
@@ -492,7 +580,7 @@ export declare const LLM_SERVICE_MODELS: {
492
580
  readonly vision: false;
493
581
  };
494
582
  };
495
- }, import("./providers").OpenAIModelNames.GPT_4_1 | import("./providers").OpenAIModelNames.GPT_4_1_MINI | import("./providers").OpenAIModelNames.GPT_5 | import("./providers").OpenAIModelNames.GPT_5_1 | import("./providers").OpenAIModelNames.GPT_5_2 | import("./providers").OpenAIModelNames.GPT_5_MINI | import("./providers").OpenAIModelNames.GPT_5_NANO>;
583
+ }, import("./providers").OpenAIModelNames.GPT_4_1 | import("./providers").OpenAIModelNames.GPT_4_1_MINI | import("./providers").OpenAIModelNames.GPT_4_1_NANO | import("./providers").OpenAIModelNames.GPT_5 | import("./providers").OpenAIModelNames.GPT_5_1 | import("./providers").OpenAIModelNames.GPT_5_2 | import("./providers").OpenAIModelNames.GPT_5_4 | import("./providers").OpenAIModelNames.GPT_5_MINI | import("./providers").OpenAIModelNames.GPT_5_NANO>;
496
584
  readonly text_to_speech: Pick<{
497
585
  readonly "gpt-4.1": {
498
586
  readonly name: import("./providers").OpenAIModelNames.GPT_4_1;
@@ -536,6 +624,27 @@ export declare const LLM_SERVICE_MODELS: {
536
624
  readonly vision: true;
537
625
  };
538
626
  };
627
+ readonly "gpt-4.1-nano": {
628
+ readonly name: import("./providers").OpenAIModelNames.GPT_4_1_NANO;
629
+ readonly limits: {
630
+ readonly maxInputTokens: 1047576;
631
+ readonly maxOutputTokens: 32768;
632
+ };
633
+ readonly config: {
634
+ readonly temperature: 0.2;
635
+ };
636
+ readonly pricing: {
637
+ readonly getPriceForTextInput: (tokens: number) => number;
638
+ readonly getPriceForTextOutput: (tokens: number) => number;
639
+ readonly getPriceForCachedTextInput: (tokens: number) => number;
640
+ readonly getPriceForAudioInput: () => number;
641
+ readonly getPriceForAudioOutput: () => number;
642
+ readonly currency: "USD";
643
+ };
644
+ readonly capabilities: {
645
+ readonly vision: true;
646
+ };
647
+ };
539
648
  readonly "gpt-5": {
540
649
  readonly name: import("./providers").OpenAIModelNames.GPT_5;
541
650
  readonly limits: {
@@ -605,6 +714,29 @@ export declare const LLM_SERVICE_MODELS: {
605
714
  readonly vision: true;
606
715
  };
607
716
  };
717
+ readonly "gpt-5.4": {
718
+ readonly name: import("./providers").OpenAIModelNames.GPT_5_4;
719
+ readonly limits: {
720
+ readonly maxInputTokens: 1050000;
721
+ readonly maxOutputTokens: 128000;
722
+ };
723
+ readonly config: {
724
+ readonly temperature: 1;
725
+ readonly reasoning_effort: "none";
726
+ readonly verbosity: "medium";
727
+ };
728
+ readonly pricing: {
729
+ readonly getPriceForTextInput: (tokens: number) => number;
730
+ readonly getPriceForTextOutput: (tokens: number) => number;
731
+ readonly getPriceForCachedTextInput: (tokens: number) => number;
732
+ readonly getPriceForAudioInput: () => number;
733
+ readonly getPriceForAudioOutput: () => number;
734
+ readonly currency: "USD";
735
+ };
736
+ readonly capabilities: {
737
+ readonly vision: true;
738
+ };
739
+ };
608
740
  readonly "gpt-5-mini": {
609
741
  readonly name: import("./providers").OpenAIModelNames.GPT_5_MINI;
610
742
  readonly limits: {
@@ -779,6 +911,27 @@ export declare const LLM_SERVICE_MODELS: {
779
911
  readonly vision: true;
780
912
  };
781
913
  };
914
+ readonly "gpt-4.1-nano": {
915
+ readonly name: import("./providers").OpenAIModelNames.GPT_4_1_NANO;
916
+ readonly limits: {
917
+ readonly maxInputTokens: 1047576;
918
+ readonly maxOutputTokens: 32768;
919
+ };
920
+ readonly config: {
921
+ readonly temperature: 0.2;
922
+ };
923
+ readonly pricing: {
924
+ readonly getPriceForTextInput: (tokens: number) => number;
925
+ readonly getPriceForTextOutput: (tokens: number) => number;
926
+ readonly getPriceForCachedTextInput: (tokens: number) => number;
927
+ readonly getPriceForAudioInput: () => number;
928
+ readonly getPriceForAudioOutput: () => number;
929
+ readonly currency: "USD";
930
+ };
931
+ readonly capabilities: {
932
+ readonly vision: true;
933
+ };
934
+ };
782
935
  readonly "gpt-5": {
783
936
  readonly name: import("./providers").OpenAIModelNames.GPT_5;
784
937
  readonly limits: {
@@ -848,6 +1001,29 @@ export declare const LLM_SERVICE_MODELS: {
848
1001
  readonly vision: true;
849
1002
  };
850
1003
  };
1004
+ readonly "gpt-5.4": {
1005
+ readonly name: import("./providers").OpenAIModelNames.GPT_5_4;
1006
+ readonly limits: {
1007
+ readonly maxInputTokens: 1050000;
1008
+ readonly maxOutputTokens: 128000;
1009
+ };
1010
+ readonly config: {
1011
+ readonly temperature: 1;
1012
+ readonly reasoning_effort: "none";
1013
+ readonly verbosity: "medium";
1014
+ };
1015
+ readonly pricing: {
1016
+ readonly getPriceForTextInput: (tokens: number) => number;
1017
+ readonly getPriceForTextOutput: (tokens: number) => number;
1018
+ readonly getPriceForCachedTextInput: (tokens: number) => number;
1019
+ readonly getPriceForAudioInput: () => number;
1020
+ readonly getPriceForAudioOutput: () => number;
1021
+ readonly currency: "USD";
1022
+ };
1023
+ readonly capabilities: {
1024
+ readonly vision: true;
1025
+ };
1026
+ };
851
1027
  readonly "gpt-5-mini": {
852
1028
  readonly name: import("./providers").OpenAIModelNames.GPT_5_MINI;
853
1029
  readonly limits: {
@@ -1088,8 +1264,55 @@ export declare const LLM_SERVICE_MODELS: {
1088
1264
  readonly vision: false;
1089
1265
  };
1090
1266
  };
1091
- readonly "gemini-3-pro-preview": {
1092
- readonly name: import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_PRO_PREVIEW;
1267
+ readonly "gemini-3-flash-preview": {
1268
+ readonly name: import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_FLASH_PREVIEW;
1269
+ readonly limits: {
1270
+ readonly maxInputTokens: 1048576;
1271
+ readonly maxOutputTokens: 65536;
1272
+ };
1273
+ readonly config: {
1274
+ readonly temperature: 1;
1275
+ readonly thinkingConfig: {
1276
+ readonly thinkingLevel: import("@google/genai").ThinkingLevel.LOW;
1277
+ };
1278
+ };
1279
+ readonly pricing: {
1280
+ readonly getPriceForTextInput: (tokens: number) => number;
1281
+ readonly getPriceForTextOutput: (tokens: number) => number;
1282
+ readonly getPriceForCachedTextInput: (tokens: number) => number;
1283
+ readonly getPriceForAudioInput: (audio_tokens: number) => number;
1284
+ readonly getPriceForCachedAudioInput: (audio_tokens: number) => number;
1285
+ readonly getPriceForAudioOutput: () => number;
1286
+ readonly currency: "USD";
1287
+ };
1288
+ readonly capabilities: {
1289
+ readonly vision: true;
1290
+ };
1291
+ };
1292
+ readonly "gemini-3.1-flash-lite-preview": {
1293
+ readonly name: import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_1_FLASH_LITE_PREVIEW;
1294
+ readonly limits: {
1295
+ readonly maxInputTokens: 1048576;
1296
+ readonly maxOutputTokens: 65536;
1297
+ };
1298
+ readonly config: {
1299
+ readonly temperature: 0.2;
1300
+ };
1301
+ readonly pricing: {
1302
+ readonly getPriceForTextInput: (tokens: number) => number;
1303
+ readonly getPriceForTextOutput: (tokens: number) => number;
1304
+ readonly getPriceForCachedTextInput: (tokens: number) => number;
1305
+ readonly getPriceForAudioInput: (audio_tokens: number) => number;
1306
+ readonly getPriceForCachedAudioInput: (audio_tokens: number) => number;
1307
+ readonly getPriceForAudioOutput: () => number;
1308
+ readonly currency: "USD";
1309
+ };
1310
+ readonly capabilities: {
1311
+ readonly vision: true;
1312
+ };
1313
+ };
1314
+ readonly "gemini-3.1-pro-preview": {
1315
+ readonly name: import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_1_PRO_PREVIEW;
1093
1316
  readonly limits: {
1094
1317
  readonly maxInputTokens: 1048576;
1095
1318
  readonly maxOutputTokens: 65536;
@@ -1113,7 +1336,7 @@ export declare const LLM_SERVICE_MODELS: {
1113
1336
  readonly vision: true;
1114
1337
  };
1115
1338
  };
1116
- }, import("./providers").GoogleGenerativeAIModelNames.GEMINI_2_5_FLASH | import("./providers").GoogleGenerativeAIModelNames.GEMINI_2_5_FLASH_LITE | import("./providers").GoogleGenerativeAIModelNames.GEMINI_2_5_PRO | import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_PRO_PREVIEW>;
1339
+ }, import("./providers").GoogleGenerativeAIModelNames.GEMINI_2_5_FLASH | import("./providers").GoogleGenerativeAIModelNames.GEMINI_2_5_FLASH_LITE | import("./providers").GoogleGenerativeAIModelNames.GEMINI_2_5_PRO | import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_FLASH_PREVIEW | import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_1_FLASH_LITE_PREVIEW | import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_1_PRO_PREVIEW>;
1117
1340
  readonly assistance: Pick<{
1118
1341
  readonly "gemini-2.5-flash": {
1119
1342
  readonly name: import("./providers").GoogleGenerativeAIModelNames.GEMINI_2_5_FLASH;
@@ -1221,8 +1444,55 @@ export declare const LLM_SERVICE_MODELS: {
1221
1444
  readonly vision: false;
1222
1445
  };
1223
1446
  };
1224
- readonly "gemini-3-pro-preview": {
1225
- readonly name: import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_PRO_PREVIEW;
1447
+ readonly "gemini-3-flash-preview": {
1448
+ readonly name: import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_FLASH_PREVIEW;
1449
+ readonly limits: {
1450
+ readonly maxInputTokens: 1048576;
1451
+ readonly maxOutputTokens: 65536;
1452
+ };
1453
+ readonly config: {
1454
+ readonly temperature: 1;
1455
+ readonly thinkingConfig: {
1456
+ readonly thinkingLevel: import("@google/genai").ThinkingLevel.LOW;
1457
+ };
1458
+ };
1459
+ readonly pricing: {
1460
+ readonly getPriceForTextInput: (tokens: number) => number;
1461
+ readonly getPriceForTextOutput: (tokens: number) => number;
1462
+ readonly getPriceForCachedTextInput: (tokens: number) => number;
1463
+ readonly getPriceForAudioInput: (audio_tokens: number) => number;
1464
+ readonly getPriceForCachedAudioInput: (audio_tokens: number) => number;
1465
+ readonly getPriceForAudioOutput: () => number;
1466
+ readonly currency: "USD";
1467
+ };
1468
+ readonly capabilities: {
1469
+ readonly vision: true;
1470
+ };
1471
+ };
1472
+ readonly "gemini-3.1-flash-lite-preview": {
1473
+ readonly name: import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_1_FLASH_LITE_PREVIEW;
1474
+ readonly limits: {
1475
+ readonly maxInputTokens: 1048576;
1476
+ readonly maxOutputTokens: 65536;
1477
+ };
1478
+ readonly config: {
1479
+ readonly temperature: 0.2;
1480
+ };
1481
+ readonly pricing: {
1482
+ readonly getPriceForTextInput: (tokens: number) => number;
1483
+ readonly getPriceForTextOutput: (tokens: number) => number;
1484
+ readonly getPriceForCachedTextInput: (tokens: number) => number;
1485
+ readonly getPriceForAudioInput: (audio_tokens: number) => number;
1486
+ readonly getPriceForCachedAudioInput: (audio_tokens: number) => number;
1487
+ readonly getPriceForAudioOutput: () => number;
1488
+ readonly currency: "USD";
1489
+ };
1490
+ readonly capabilities: {
1491
+ readonly vision: true;
1492
+ };
1493
+ };
1494
+ readonly "gemini-3.1-pro-preview": {
1495
+ readonly name: import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_1_PRO_PREVIEW;
1226
1496
  readonly limits: {
1227
1497
  readonly maxInputTokens: 1048576;
1228
1498
  readonly maxOutputTokens: 65536;
@@ -1246,7 +1516,7 @@ export declare const LLM_SERVICE_MODELS: {
1246
1516
  readonly vision: true;
1247
1517
  };
1248
1518
  };
1249
- }, import("./providers").GoogleGenerativeAIModelNames.GEMINI_2_5_FLASH | import("./providers").GoogleGenerativeAIModelNames.GEMINI_2_5_FLASH_LITE | import("./providers").GoogleGenerativeAIModelNames.GEMINI_2_5_PRO | import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_PRO_PREVIEW>;
1519
+ }, import("./providers").GoogleGenerativeAIModelNames.GEMINI_2_5_FLASH | import("./providers").GoogleGenerativeAIModelNames.GEMINI_2_5_FLASH_LITE | import("./providers").GoogleGenerativeAIModelNames.GEMINI_2_5_PRO | import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_FLASH_PREVIEW | import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_1_FLASH_LITE_PREVIEW | import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_1_PRO_PREVIEW>;
1250
1520
  readonly speech_to_text: Pick<{
1251
1521
  readonly "gemini-2.5-flash": {
1252
1522
  readonly name: import("./providers").GoogleGenerativeAIModelNames.GEMINI_2_5_FLASH;
@@ -1354,8 +1624,55 @@ export declare const LLM_SERVICE_MODELS: {
1354
1624
  readonly vision: false;
1355
1625
  };
1356
1626
  };
1357
- readonly "gemini-3-pro-preview": {
1358
- readonly name: import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_PRO_PREVIEW;
1627
+ readonly "gemini-3-flash-preview": {
1628
+ readonly name: import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_FLASH_PREVIEW;
1629
+ readonly limits: {
1630
+ readonly maxInputTokens: 1048576;
1631
+ readonly maxOutputTokens: 65536;
1632
+ };
1633
+ readonly config: {
1634
+ readonly temperature: 1;
1635
+ readonly thinkingConfig: {
1636
+ readonly thinkingLevel: import("@google/genai").ThinkingLevel.LOW;
1637
+ };
1638
+ };
1639
+ readonly pricing: {
1640
+ readonly getPriceForTextInput: (tokens: number) => number;
1641
+ readonly getPriceForTextOutput: (tokens: number) => number;
1642
+ readonly getPriceForCachedTextInput: (tokens: number) => number;
1643
+ readonly getPriceForAudioInput: (audio_tokens: number) => number;
1644
+ readonly getPriceForCachedAudioInput: (audio_tokens: number) => number;
1645
+ readonly getPriceForAudioOutput: () => number;
1646
+ readonly currency: "USD";
1647
+ };
1648
+ readonly capabilities: {
1649
+ readonly vision: true;
1650
+ };
1651
+ };
1652
+ readonly "gemini-3.1-flash-lite-preview": {
1653
+ readonly name: import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_1_FLASH_LITE_PREVIEW;
1654
+ readonly limits: {
1655
+ readonly maxInputTokens: 1048576;
1656
+ readonly maxOutputTokens: 65536;
1657
+ };
1658
+ readonly config: {
1659
+ readonly temperature: 0.2;
1660
+ };
1661
+ readonly pricing: {
1662
+ readonly getPriceForTextInput: (tokens: number) => number;
1663
+ readonly getPriceForTextOutput: (tokens: number) => number;
1664
+ readonly getPriceForCachedTextInput: (tokens: number) => number;
1665
+ readonly getPriceForAudioInput: (audio_tokens: number) => number;
1666
+ readonly getPriceForCachedAudioInput: (audio_tokens: number) => number;
1667
+ readonly getPriceForAudioOutput: () => number;
1668
+ readonly currency: "USD";
1669
+ };
1670
+ readonly capabilities: {
1671
+ readonly vision: true;
1672
+ };
1673
+ };
1674
+ readonly "gemini-3.1-pro-preview": {
1675
+ readonly name: import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_1_PRO_PREVIEW;
1359
1676
  readonly limits: {
1360
1677
  readonly maxInputTokens: 1048576;
1361
1678
  readonly maxOutputTokens: 65536;
@@ -1379,7 +1696,7 @@ export declare const LLM_SERVICE_MODELS: {
1379
1696
  readonly vision: true;
1380
1697
  };
1381
1698
  };
1382
- }, import("./providers").GoogleGenerativeAIModelNames.GEMINI_2_5_FLASH | import("./providers").GoogleGenerativeAIModelNames.GEMINI_2_5_FLASH_LITE | import("./providers").GoogleGenerativeAIModelNames.GEMINI_2_5_PRO | import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_PRO_PREVIEW>;
1699
+ }, import("./providers").GoogleGenerativeAIModelNames.GEMINI_2_5_FLASH | import("./providers").GoogleGenerativeAIModelNames.GEMINI_2_5_FLASH_LITE | import("./providers").GoogleGenerativeAIModelNames.GEMINI_2_5_PRO | import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_FLASH_PREVIEW | import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_1_FLASH_LITE_PREVIEW | import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_1_PRO_PREVIEW>;
1383
1700
  readonly text_to_speech: Pick<{
1384
1701
  readonly "gemini-2.5-flash": {
1385
1702
  readonly name: import("./providers").GoogleGenerativeAIModelNames.GEMINI_2_5_FLASH;
@@ -1487,8 +1804,55 @@ export declare const LLM_SERVICE_MODELS: {
1487
1804
  readonly vision: false;
1488
1805
  };
1489
1806
  };
1490
- readonly "gemini-3-pro-preview": {
1491
- readonly name: import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_PRO_PREVIEW;
1807
+ readonly "gemini-3-flash-preview": {
1808
+ readonly name: import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_FLASH_PREVIEW;
1809
+ readonly limits: {
1810
+ readonly maxInputTokens: 1048576;
1811
+ readonly maxOutputTokens: 65536;
1812
+ };
1813
+ readonly config: {
1814
+ readonly temperature: 1;
1815
+ readonly thinkingConfig: {
1816
+ readonly thinkingLevel: import("@google/genai").ThinkingLevel.LOW;
1817
+ };
1818
+ };
1819
+ readonly pricing: {
1820
+ readonly getPriceForTextInput: (tokens: number) => number;
1821
+ readonly getPriceForTextOutput: (tokens: number) => number;
1822
+ readonly getPriceForCachedTextInput: (tokens: number) => number;
1823
+ readonly getPriceForAudioInput: (audio_tokens: number) => number;
1824
+ readonly getPriceForCachedAudioInput: (audio_tokens: number) => number;
1825
+ readonly getPriceForAudioOutput: () => number;
1826
+ readonly currency: "USD";
1827
+ };
1828
+ readonly capabilities: {
1829
+ readonly vision: true;
1830
+ };
1831
+ };
1832
+ readonly "gemini-3.1-flash-lite-preview": {
1833
+ readonly name: import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_1_FLASH_LITE_PREVIEW;
1834
+ readonly limits: {
1835
+ readonly maxInputTokens: 1048576;
1836
+ readonly maxOutputTokens: 65536;
1837
+ };
1838
+ readonly config: {
1839
+ readonly temperature: 0.2;
1840
+ };
1841
+ readonly pricing: {
1842
+ readonly getPriceForTextInput: (tokens: number) => number;
1843
+ readonly getPriceForTextOutput: (tokens: number) => number;
1844
+ readonly getPriceForCachedTextInput: (tokens: number) => number;
1845
+ readonly getPriceForAudioInput: (audio_tokens: number) => number;
1846
+ readonly getPriceForCachedAudioInput: (audio_tokens: number) => number;
1847
+ readonly getPriceForAudioOutput: () => number;
1848
+ readonly currency: "USD";
1849
+ };
1850
+ readonly capabilities: {
1851
+ readonly vision: true;
1852
+ };
1853
+ };
1854
+ readonly "gemini-3.1-pro-preview": {
1855
+ readonly name: import("./providers").GoogleGenerativeAIModelNames.GEMINI_3_1_PRO_PREVIEW;
1492
1856
  readonly limits: {
1493
1857
  readonly maxInputTokens: 1048576;
1494
1858
  readonly maxOutputTokens: 65536;