llm.rb 12.4.0 → 12.5.0

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.
data/data/google.json CHANGED
@@ -9,52 +9,38 @@
9
9
  "name": "Google",
10
10
  "doc": "https://ai.google.dev/gemini-api/docs/models",
11
11
  "models": {
12
- "gemini-3.1-flash-lite": {
13
- "id": "gemini-3.1-flash-lite",
14
- "name": "Gemini 3.1 Flash Lite",
15
- "description": "Low-latency Gemini model for high-volume multimodal and agent workloads",
16
- "family": "gemini-flash-lite",
12
+ "gemini-2.5-flash-image": {
13
+ "id": "gemini-2.5-flash-image",
14
+ "name": "Nano Banana",
15
+ "description": "Nano Banana image model for fast generation, edits, and character-consistent assets",
16
+ "family": "gemini-flash",
17
17
  "attachment": true,
18
18
  "reasoning": true,
19
- "reasoning_options": [
20
- {
21
- "type": "effort",
22
- "values": [
23
- "minimal",
24
- "low",
25
- "medium",
26
- "high"
27
- ]
28
- }
29
- ],
30
- "tool_call": true,
31
- "structured_output": true,
19
+ "reasoning_options": [],
20
+ "tool_call": false,
32
21
  "temperature": true,
33
- "knowledge": "2025-01",
34
- "release_date": "2026-05-07",
35
- "last_updated": "2026-05-07",
22
+ "knowledge": "2024-06",
23
+ "release_date": "2025-08-26",
24
+ "last_updated": "2025-08-26",
36
25
  "modalities": {
37
26
  "input": [
38
27
  "text",
39
- "image",
40
- "video",
41
- "audio",
42
- "pdf"
28
+ "image"
43
29
  ],
44
30
  "output": [
45
- "text"
31
+ "text",
32
+ "image"
46
33
  ]
47
34
  },
48
35
  "open_weights": false,
49
36
  "limit": {
50
- "context": 1048576,
51
- "output": 65536
37
+ "context": 32768,
38
+ "output": 32768
52
39
  },
53
40
  "cost": {
54
- "input": 0.25,
55
- "output": 1.5,
56
- "cache_read": 0.025,
57
- "input_audio": 0.5
41
+ "input": 0.3,
42
+ "output": 30,
43
+ "cache_read": 0.075
58
44
  }
59
45
  },
60
46
  "gemini-2.5-flash-preview-tts": {
@@ -87,94 +73,36 @@
87
73
  "output": 10
88
74
  }
89
75
  },
90
- "gemini-2.5-pro": {
91
- "id": "gemini-2.5-pro",
92
- "name": "Gemini 2.5 Pro",
93
- "description": "Google's proven reasoning model for coding, math, and multimodal analysis",
94
- "family": "gemini-pro",
76
+ "gemini-3.1-flash-lite": {
77
+ "id": "gemini-3.1-flash-lite",
78
+ "name": "Gemini 3.1 Flash Lite",
79
+ "description": "Low-latency Gemini model for high-volume multimodal and agent workloads",
80
+ "family": "gemini-flash-lite",
95
81
  "attachment": true,
96
82
  "reasoning": true,
97
83
  "reasoning_options": [
98
84
  {
99
- "type": "budget_tokens",
100
- "min": 128,
101
- "max": 32768
85
+ "type": "effort",
86
+ "values": [
87
+ "minimal",
88
+ "low",
89
+ "medium",
90
+ "high"
91
+ ]
102
92
  }
103
93
  ],
104
94
  "tool_call": true,
105
95
  "structured_output": true,
106
96
  "temperature": true,
107
97
  "knowledge": "2025-01",
108
- "release_date": "2025-06-17",
109
- "last_updated": "2025-06-17",
98
+ "release_date": "2026-05-07",
99
+ "last_updated": "2026-05-07",
110
100
  "modalities": {
111
101
  "input": [
112
102
  "text",
113
103
  "image",
114
- "audio",
115
104
  "video",
116
- "pdf"
117
- ],
118
- "output": [
119
- "text"
120
- ]
121
- },
122
- "open_weights": false,
123
- "limit": {
124
- "context": 1048576,
125
- "output": 65536
126
- },
127
- "cost": {
128
- "input": 1.25,
129
- "output": 10,
130
- "cache_read": 0.125,
131
- "tiers": [
132
- {
133
- "input": 2.5,
134
- "output": 15,
135
- "cache_read": 0.25,
136
- "tier": {
137
- "type": "context",
138
- "size": 200000
139
- }
140
- }
141
- ],
142
- "context_over_200k": {
143
- "input": 2.5,
144
- "output": 15,
145
- "cache_read": 0.25
146
- }
147
- }
148
- },
149
- "gemini-2.5-flash": {
150
- "id": "gemini-2.5-flash",
151
- "name": "Gemini 2.5 Flash",
152
- "description": "Fast Gemini workhorse for multimodal apps where latency and price matter",
153
- "family": "gemini-flash",
154
- "attachment": true,
155
- "reasoning": true,
156
- "reasoning_options": [
157
- {
158
- "type": "toggle"
159
- },
160
- {
161
- "type": "budget_tokens",
162
- "min": 0,
163
- "max": 24576
164
- }
165
- ],
166
- "tool_call": true,
167
- "structured_output": true,
168
- "temperature": true,
169
- "knowledge": "2025-01",
170
- "release_date": "2025-06-17",
171
- "last_updated": "2025-06-17",
172
- "modalities": {
173
- "input": [
174
- "text",
175
- "image",
176
105
  "audio",
177
- "video",
178
106
  "pdf"
179
107
  ],
180
108
  "output": [
@@ -187,10 +115,10 @@
187
115
  "output": 65536
188
116
  },
189
117
  "cost": {
190
- "input": 0.3,
191
- "output": 2.5,
192
- "cache_read": 0.03,
193
- "input_audio": 1
118
+ "input": 0.25,
119
+ "output": 1.5,
120
+ "cache_read": 0.025,
121
+ "input_audio": 0.5
194
122
  }
195
123
  },
196
124
  "gemini-3.5-flash": {
@@ -241,10 +169,10 @@
241
169
  "input_audio": 1.5
242
170
  }
243
171
  },
244
- "gemma-4-31b-it": {
245
- "id": "gemma-4-31b-it",
246
- "name": "Gemma 4 31B IT",
247
- "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning",
172
+ "gemma-4-26b-a4b-it": {
173
+ "id": "gemma-4-26b-a4b-it",
174
+ "name": "Gemma 4 26B A4B IT",
175
+ "description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
248
176
  "family": "gemma",
249
177
  "attachment": true,
250
178
  "reasoning": true,
@@ -310,40 +238,10 @@
310
238
  "cache_read": 0.025
311
239
  }
312
240
  },
313
- "gemini-embedding-001": {
314
- "id": "gemini-embedding-001",
315
- "name": "Gemini Embedding 001",
316
- "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines",
317
- "family": "gemini",
318
- "attachment": false,
319
- "reasoning": false,
320
- "tool_call": false,
321
- "temperature": false,
322
- "knowledge": "2025-05",
323
- "release_date": "2025-05-20",
324
- "last_updated": "2025-05-20",
325
- "modalities": {
326
- "input": [
327
- "text"
328
- ],
329
- "output": [
330
- "text"
331
- ]
332
- },
333
- "open_weights": false,
334
- "limit": {
335
- "context": 2048,
336
- "output": 1
337
- },
338
- "cost": {
339
- "input": 0.15,
340
- "output": 0
341
- }
342
- },
343
- "gemini-3.1-pro-preview-customtools": {
344
- "id": "gemini-3.1-pro-preview-customtools",
345
- "name": "Gemini 3.1 Pro Preview Custom Tools",
346
- "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis",
241
+ "gemini-3-pro-preview": {
242
+ "id": "gemini-3-pro-preview",
243
+ "name": "Gemini 3 Pro Preview",
244
+ "description": "Preview Gemini flagship for complex reasoning, coding, and rich multimodal prompts",
347
245
  "family": "gemini-pro",
348
246
  "attachment": true,
349
247
  "reasoning": true,
@@ -352,7 +250,6 @@
352
250
  "type": "effort",
353
251
  "values": [
354
252
  "low",
355
- "medium",
356
253
  "high"
357
254
  ]
358
255
  }
@@ -361,8 +258,8 @@
361
258
  "structured_output": true,
362
259
  "temperature": true,
363
260
  "knowledge": "2025-01",
364
- "release_date": "2026-02-19",
365
- "last_updated": "2026-02-19",
261
+ "release_date": "2025-11-18",
262
+ "last_updated": "2025-11-18",
366
263
  "modalities": {
367
264
  "input": [
368
265
  "text",
@@ -380,6 +277,7 @@
380
277
  "context": 1048576,
381
278
  "output": 65536
382
279
  },
280
+ "status": "deprecated",
383
281
  "cost": {
384
282
  "input": 2,
385
283
  "output": 12,
@@ -402,35 +300,36 @@
402
300
  }
403
301
  }
404
302
  },
405
- "gemini-flash-lite-latest": {
406
- "id": "gemini-flash-lite-latest",
407
- "name": "Gemini Flash-Lite Latest",
408
- "description": "Low-latency Gemini model for high-volume multimodal and agent workloads",
303
+ "gemini-3.1-flash-lite-preview": {
304
+ "id": "gemini-3.1-flash-lite-preview",
305
+ "name": "Gemini 3.1 Flash Lite Preview",
306
+ "description": "Legacy model retained for compatibility with older integrations",
409
307
  "family": "gemini-flash-lite",
410
308
  "attachment": true,
411
309
  "reasoning": true,
412
310
  "reasoning_options": [
413
311
  {
414
- "type": "toggle"
415
- },
416
- {
417
- "type": "budget_tokens",
418
- "min": 512,
419
- "max": 24576
312
+ "type": "effort",
313
+ "values": [
314
+ "minimal",
315
+ "low",
316
+ "medium",
317
+ "high"
318
+ ]
420
319
  }
421
320
  ],
422
321
  "tool_call": true,
423
322
  "structured_output": true,
424
323
  "temperature": true,
425
324
  "knowledge": "2025-01",
426
- "release_date": "2025-09-25",
427
- "last_updated": "2025-09-25",
325
+ "release_date": "2026-03-03",
326
+ "last_updated": "2026-03-03",
428
327
  "modalities": {
429
328
  "input": [
430
329
  "text",
431
330
  "image",
432
- "audio",
433
331
  "video",
332
+ "audio",
434
333
  "pdf"
435
334
  ],
436
335
  "output": [
@@ -442,10 +341,44 @@
442
341
  "context": 1048576,
443
342
  "output": 65536
444
343
  },
344
+ "status": "deprecated",
445
345
  "cost": {
446
- "input": 0.1,
447
- "output": 0.4,
448
- "cache_read": 0.025
346
+ "input": 0.25,
347
+ "output": 1.5,
348
+ "cache_read": 0.025,
349
+ "input_audio": 0.5
350
+ }
351
+ },
352
+ "gemma-4-31b-it": {
353
+ "id": "gemma-4-31b-it",
354
+ "name": "Gemma 4 31B IT",
355
+ "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning",
356
+ "family": "gemma",
357
+ "attachment": true,
358
+ "reasoning": true,
359
+ "reasoning_options": [
360
+ {
361
+ "type": "toggle"
362
+ }
363
+ ],
364
+ "tool_call": true,
365
+ "structured_output": true,
366
+ "temperature": true,
367
+ "release_date": "2026-04-02",
368
+ "last_updated": "2026-04-02",
369
+ "modalities": {
370
+ "input": [
371
+ "text",
372
+ "image"
373
+ ],
374
+ "output": [
375
+ "text"
376
+ ]
377
+ },
378
+ "open_weights": true,
379
+ "limit": {
380
+ "context": 262144,
381
+ "output": 32768
449
382
  }
450
383
  },
451
384
  "gemini-3-pro-image-preview": {
@@ -481,55 +414,161 @@
481
414
  "output": 120
482
415
  }
483
416
  },
484
- "gemini-2.5-flash-image": {
485
- "id": "gemini-2.5-flash-image",
486
- "name": "Nano Banana",
487
- "description": "Nano Banana image model for fast generation, edits, and character-consistent assets",
488
- "family": "gemini-flash",
489
- "attachment": true,
490
- "reasoning": true,
491
- "reasoning_options": [],
417
+ "gemini-embedding-001": {
418
+ "id": "gemini-embedding-001",
419
+ "name": "Gemini Embedding 001",
420
+ "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines",
421
+ "family": "gemini",
422
+ "attachment": false,
423
+ "reasoning": false,
492
424
  "tool_call": false,
425
+ "temperature": false,
426
+ "knowledge": "2025-05",
427
+ "release_date": "2025-05-20",
428
+ "last_updated": "2025-05-20",
429
+ "modalities": {
430
+ "input": [
431
+ "text"
432
+ ],
433
+ "output": [
434
+ "text"
435
+ ]
436
+ },
437
+ "open_weights": false,
438
+ "limit": {
439
+ "context": 2048,
440
+ "output": 1
441
+ },
442
+ "cost": {
443
+ "input": 0.15,
444
+ "output": 0
445
+ }
446
+ },
447
+ "gemini-2.0-flash-lite": {
448
+ "id": "gemini-2.0-flash-lite",
449
+ "name": "Gemini 2.0 Flash-Lite",
450
+ "description": "Legacy model retained for compatibility with older integrations",
451
+ "family": "gemini-flash-lite",
452
+ "attachment": true,
453
+ "reasoning": false,
454
+ "tool_call": true,
455
+ "structured_output": true,
493
456
  "temperature": true,
494
457
  "knowledge": "2024-06",
495
- "release_date": "2025-08-26",
496
- "last_updated": "2025-08-26",
458
+ "release_date": "2024-12-11",
459
+ "last_updated": "2024-12-11",
497
460
  "modalities": {
498
461
  "input": [
499
462
  "text",
500
- "image"
463
+ "image",
464
+ "audio",
465
+ "video",
466
+ "pdf"
467
+ ],
468
+ "output": [
469
+ "text"
470
+ ]
471
+ },
472
+ "open_weights": false,
473
+ "limit": {
474
+ "context": 1048576,
475
+ "output": 8192
476
+ },
477
+ "status": "deprecated",
478
+ "cost": {
479
+ "input": 0.075,
480
+ "output": 0.3
481
+ }
482
+ },
483
+ "gemini-2.5-pro-preview-tts": {
484
+ "id": "gemini-2.5-pro-preview-tts",
485
+ "name": "Gemini 2.5 Pro Preview TTS",
486
+ "description": "Speech generation model for controllable voice, narration, and audio delivery",
487
+ "family": "gemini-flash",
488
+ "attachment": false,
489
+ "reasoning": false,
490
+ "tool_call": false,
491
+ "temperature": true,
492
+ "knowledge": "2025-01",
493
+ "release_date": "2025-05-01",
494
+ "last_updated": "2025-05-01",
495
+ "modalities": {
496
+ "input": [
497
+ "text"
501
498
  ],
502
499
  "output": [
500
+ "audio"
501
+ ]
502
+ },
503
+ "open_weights": false,
504
+ "limit": {
505
+ "context": 8192,
506
+ "output": 16384
507
+ },
508
+ "cost": {
509
+ "input": 1,
510
+ "output": 20
511
+ }
512
+ },
513
+ "gemini-2.5-flash-lite": {
514
+ "id": "gemini-2.5-flash-lite",
515
+ "name": "Gemini 2.5 Flash-Lite",
516
+ "description": "Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents",
517
+ "family": "gemini-flash-lite",
518
+ "attachment": true,
519
+ "reasoning": true,
520
+ "reasoning_options": [
521
+ {
522
+ "type": "toggle"
523
+ },
524
+ {
525
+ "type": "budget_tokens",
526
+ "min": 512,
527
+ "max": 24576
528
+ }
529
+ ],
530
+ "tool_call": true,
531
+ "structured_output": true,
532
+ "temperature": true,
533
+ "knowledge": "2025-01",
534
+ "release_date": "2025-06-17",
535
+ "last_updated": "2025-06-17",
536
+ "modalities": {
537
+ "input": [
503
538
  "text",
504
- "image"
539
+ "image",
540
+ "audio",
541
+ "video",
542
+ "pdf"
543
+ ],
544
+ "output": [
545
+ "text"
505
546
  ]
506
547
  },
507
548
  "open_weights": false,
508
549
  "limit": {
509
- "context": 32768,
510
- "output": 32768
550
+ "context": 1048576,
551
+ "output": 65536
511
552
  },
512
553
  "cost": {
513
- "input": 0.3,
514
- "output": 30,
515
- "cache_read": 0.075
554
+ "input": 0.1,
555
+ "output": 0.4,
556
+ "cache_read": 0.01,
557
+ "input_audio": 0.3
516
558
  }
517
559
  },
518
- "gemini-2.5-flash-lite": {
519
- "id": "gemini-2.5-flash-lite",
520
- "name": "Gemini 2.5 Flash-Lite",
521
- "description": "Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents",
522
- "family": "gemini-flash-lite",
560
+ "gemini-2.5-pro": {
561
+ "id": "gemini-2.5-pro",
562
+ "name": "Gemini 2.5 Pro",
563
+ "description": "Google's proven reasoning model for coding, math, and multimodal analysis",
564
+ "family": "gemini-pro",
523
565
  "attachment": true,
524
566
  "reasoning": true,
525
567
  "reasoning_options": [
526
- {
527
- "type": "toggle"
528
- },
529
568
  {
530
569
  "type": "budget_tokens",
531
- "min": 512,
532
- "max": 24576
570
+ "min": 128,
571
+ "max": 32768
533
572
  }
534
573
  ],
535
574
  "tool_call": true,
@@ -556,10 +595,25 @@
556
595
  "output": 65536
557
596
  },
558
597
  "cost": {
559
- "input": 0.1,
560
- "output": 0.4,
561
- "cache_read": 0.01,
562
- "input_audio": 0.3
598
+ "input": 1.25,
599
+ "output": 10,
600
+ "cache_read": 0.125,
601
+ "tiers": [
602
+ {
603
+ "input": 2.5,
604
+ "output": 15,
605
+ "cache_read": 0.25,
606
+ "tier": {
607
+ "type": "context",
608
+ "size": 200000
609
+ }
610
+ }
611
+ ],
612
+ "context_over_200k": {
613
+ "input": 2.5,
614
+ "output": 15,
615
+ "cache_read": 0.25
616
+ }
563
617
  }
564
618
  },
565
619
  "gemini-omni-flash-preview": {
@@ -636,17 +690,18 @@
636
690
  "output": 60
637
691
  }
638
692
  },
639
- "gemini-3.1-pro-preview": {
640
- "id": "gemini-3.1-pro-preview",
641
- "name": "Gemini 3.1 Pro Preview",
642
- "description": "Reasoning-first Gemini preview for agentic coding and complex problem solving",
643
- "family": "gemini-pro",
693
+ "gemini-3-flash-preview": {
694
+ "id": "gemini-3-flash-preview",
695
+ "name": "Gemini 3 Flash Preview",
696
+ "description": "New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs",
697
+ "family": "gemini-flash",
644
698
  "attachment": true,
645
699
  "reasoning": true,
646
700
  "reasoning_options": [
647
701
  {
648
702
  "type": "effort",
649
703
  "values": [
704
+ "minimal",
650
705
  "low",
651
706
  "medium",
652
707
  "high"
@@ -657,8 +712,8 @@
657
712
  "structured_output": true,
658
713
  "temperature": true,
659
714
  "knowledge": "2025-01",
660
- "release_date": "2026-02-19",
661
- "last_updated": "2026-02-19",
715
+ "release_date": "2025-12-17",
716
+ "last_updated": "2025-12-17",
662
717
  "modalities": {
663
718
  "input": [
664
719
  "text",
@@ -677,63 +732,16 @@
677
732
  "output": 65536
678
733
  },
679
734
  "cost": {
680
- "input": 2,
681
- "output": 12,
682
- "cache_read": 0.2,
683
- "tiers": [
684
- {
685
- "input": 4,
686
- "output": 18,
687
- "cache_read": 0.4,
688
- "tier": {
689
- "type": "context",
690
- "size": 200000
691
- }
692
- }
693
- ],
694
- "context_over_200k": {
695
- "input": 4,
696
- "output": 18,
697
- "cache_read": 0.4
698
- }
699
- }
700
- },
701
- "gemma-4-26b-a4b-it": {
702
- "id": "gemma-4-26b-a4b-it",
703
- "name": "Gemma 4 26B A4B IT",
704
- "description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
705
- "family": "gemma",
706
- "attachment": true,
707
- "reasoning": true,
708
- "reasoning_options": [
709
- {
710
- "type": "toggle"
711
- }
712
- ],
713
- "tool_call": true,
714
- "structured_output": true,
715
- "temperature": true,
716
- "release_date": "2026-04-02",
717
- "last_updated": "2026-04-02",
718
- "modalities": {
719
- "input": [
720
- "text",
721
- "image"
722
- ],
723
- "output": [
724
- "text"
725
- ]
726
- },
727
- "open_weights": true,
728
- "limit": {
729
- "context": 262144,
730
- "output": 32768
735
+ "input": 0.5,
736
+ "output": 3,
737
+ "cache_read": 0.05,
738
+ "input_audio": 1
731
739
  }
732
740
  },
733
- "gemini-3-pro-preview": {
734
- "id": "gemini-3-pro-preview",
735
- "name": "Gemini 3 Pro Preview",
736
- "description": "Preview Gemini flagship for complex reasoning, coding, and rich multimodal prompts",
741
+ "gemini-3.1-pro-preview-customtools": {
742
+ "id": "gemini-3.1-pro-preview-customtools",
743
+ "name": "Gemini 3.1 Pro Preview Custom Tools",
744
+ "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis",
737
745
  "family": "gemini-pro",
738
746
  "attachment": true,
739
747
  "reasoning": true,
@@ -742,6 +750,7 @@
742
750
  "type": "effort",
743
751
  "values": [
744
752
  "low",
753
+ "medium",
745
754
  "high"
746
755
  ]
747
756
  }
@@ -750,8 +759,8 @@
750
759
  "structured_output": true,
751
760
  "temperature": true,
752
761
  "knowledge": "2025-01",
753
- "release_date": "2025-11-18",
754
- "last_updated": "2025-11-18",
762
+ "release_date": "2026-02-19",
763
+ "last_updated": "2026-02-19",
755
764
  "modalities": {
756
765
  "input": [
757
766
  "text",
@@ -769,7 +778,6 @@
769
778
  "context": 1048576,
770
779
  "output": 65536
771
780
  },
772
- "status": "deprecated",
773
781
  "cost": {
774
782
  "input": 2,
775
783
  "output": 12,
@@ -792,10 +800,10 @@
792
800
  }
793
801
  }
794
802
  },
795
- "gemini-3-flash-preview": {
796
- "id": "gemini-3-flash-preview",
797
- "name": "Gemini 3 Flash Preview",
798
- "description": "New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs",
803
+ "gemini-flash-latest": {
804
+ "id": "gemini-flash-latest",
805
+ "name": "Gemini Flash Latest",
806
+ "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost",
799
807
  "family": "gemini-flash",
800
808
  "attachment": true,
801
809
  "reasoning": true,
@@ -814,8 +822,8 @@
814
822
  "structured_output": true,
815
823
  "temperature": true,
816
824
  "knowledge": "2025-01",
817
- "release_date": "2025-12-17",
818
- "last_updated": "2025-12-17",
825
+ "release_date": "2026-05-19",
826
+ "last_updated": "2026-05-19",
819
827
  "modalities": {
820
828
  "input": [
821
829
  "text",
@@ -834,46 +842,16 @@
834
842
  "output": 65536
835
843
  },
836
844
  "cost": {
837
- "input": 0.5,
838
- "output": 3,
839
- "cache_read": 0.05,
840
- "input_audio": 1
841
- }
842
- },
843
- "gemini-2.5-pro-preview-tts": {
844
- "id": "gemini-2.5-pro-preview-tts",
845
- "name": "Gemini 2.5 Pro Preview TTS",
846
- "description": "Speech generation model for controllable voice, narration, and audio delivery",
847
- "family": "gemini-flash",
848
- "attachment": false,
849
- "reasoning": false,
850
- "tool_call": false,
851
- "temperature": true,
852
- "knowledge": "2025-01",
853
- "release_date": "2025-05-01",
854
- "last_updated": "2025-05-01",
855
- "modalities": {
856
- "input": [
857
- "text"
858
- ],
859
- "output": [
860
- "audio"
861
- ]
862
- },
863
- "open_weights": false,
864
- "limit": {
865
- "context": 8192,
866
- "output": 16384
867
- },
868
- "cost": {
869
- "input": 1,
870
- "output": 20
845
+ "input": 1.5,
846
+ "output": 9,
847
+ "cache_read": 0.15,
848
+ "input_audio": 1.5
871
849
  }
872
850
  },
873
- "gemini-flash-latest": {
874
- "id": "gemini-flash-latest",
875
- "name": "Gemini Flash Latest",
876
- "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost",
851
+ "gemini-2.5-flash": {
852
+ "id": "gemini-2.5-flash",
853
+ "name": "Gemini 2.5 Flash",
854
+ "description": "Fast Gemini workhorse for multimodal apps where latency and price matter",
877
855
  "family": "gemini-flash",
878
856
  "attachment": true,
879
857
  "reasoning": true,
@@ -891,8 +869,8 @@
891
869
  "structured_output": true,
892
870
  "temperature": true,
893
871
  "knowledge": "2025-01",
894
- "release_date": "2025-09-25",
895
- "last_updated": "2025-09-25",
872
+ "release_date": "2025-06-17",
873
+ "last_updated": "2025-06-17",
896
874
  "modalities": {
897
875
  "input": [
898
876
  "text",
@@ -913,14 +891,14 @@
913
891
  "cost": {
914
892
  "input": 0.3,
915
893
  "output": 2.5,
916
- "cache_read": 0.075,
894
+ "cache_read": 0.03,
917
895
  "input_audio": 1
918
896
  }
919
897
  },
920
- "gemini-3.1-flash-lite-preview": {
921
- "id": "gemini-3.1-flash-lite-preview",
922
- "name": "Gemini 3.1 Flash Lite Preview",
923
- "description": "Legacy model retained for compatibility with older integrations",
898
+ "gemini-flash-lite-latest": {
899
+ "id": "gemini-flash-lite-latest",
900
+ "name": "Gemini Flash-Lite Latest",
901
+ "description": "Low-latency Gemini model for high-volume multimodal and agent workloads",
924
902
  "family": "gemini-flash-lite",
925
903
  "attachment": true,
926
904
  "reasoning": true,
@@ -939,8 +917,8 @@
939
917
  "structured_output": true,
940
918
  "temperature": true,
941
919
  "knowledge": "2025-01",
942
- "release_date": "2026-03-03",
943
- "last_updated": "2026-03-03",
920
+ "release_date": "2026-05-07",
921
+ "last_updated": "2026-05-07",
944
922
  "modalities": {
945
923
  "input": [
946
924
  "text",
@@ -958,7 +936,6 @@
958
936
  "context": 1048576,
959
937
  "output": 65536
960
938
  },
961
- "status": "deprecated",
962
939
  "cost": {
963
940
  "input": 0.25,
964
941
  "output": 1.5,
@@ -966,25 +943,35 @@
966
943
  "input_audio": 0.5
967
944
  }
968
945
  },
969
- "gemini-2.0-flash-lite": {
970
- "id": "gemini-2.0-flash-lite",
971
- "name": "Gemini 2.0 Flash-Lite",
972
- "description": "Legacy model retained for compatibility with older integrations",
973
- "family": "gemini-flash-lite",
946
+ "gemini-3.1-pro-preview": {
947
+ "id": "gemini-3.1-pro-preview",
948
+ "name": "Gemini 3.1 Pro Preview",
949
+ "description": "Reasoning-first Gemini preview for agentic coding and complex problem solving",
950
+ "family": "gemini-pro",
974
951
  "attachment": true,
975
- "reasoning": false,
952
+ "reasoning": true,
953
+ "reasoning_options": [
954
+ {
955
+ "type": "effort",
956
+ "values": [
957
+ "low",
958
+ "medium",
959
+ "high"
960
+ ]
961
+ }
962
+ ],
976
963
  "tool_call": true,
977
964
  "structured_output": true,
978
965
  "temperature": true,
979
- "knowledge": "2024-06",
980
- "release_date": "2024-12-11",
981
- "last_updated": "2024-12-11",
966
+ "knowledge": "2025-01",
967
+ "release_date": "2026-02-19",
968
+ "last_updated": "2026-02-19",
982
969
  "modalities": {
983
970
  "input": [
984
971
  "text",
985
972
  "image",
986
- "audio",
987
973
  "video",
974
+ "audio",
988
975
  "pdf"
989
976
  ],
990
977
  "output": [
@@ -994,12 +981,28 @@
994
981
  "open_weights": false,
995
982
  "limit": {
996
983
  "context": 1048576,
997
- "output": 8192
984
+ "output": 65536
998
985
  },
999
- "status": "deprecated",
1000
986
  "cost": {
1001
- "input": 0.075,
1002
- "output": 0.3
987
+ "input": 2,
988
+ "output": 12,
989
+ "cache_read": 0.2,
990
+ "tiers": [
991
+ {
992
+ "input": 4,
993
+ "output": 18,
994
+ "cache_read": 0.4,
995
+ "tier": {
996
+ "type": "context",
997
+ "size": 200000
998
+ }
999
+ }
1000
+ ],
1001
+ "context_over_200k": {
1002
+ "input": 4,
1003
+ "output": 18,
1004
+ "cache_read": 0.4
1005
+ }
1003
1006
  }
1004
1007
  }
1005
1008
  }