model-library 0.1.3__py3-none-any.whl → 0.1.5__py3-none-any.whl

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.
Files changed (37) hide show
  1. model_library/base/base.py +2 -0
  2. model_library/base/output.py +24 -9
  3. model_library/base/utils.py +27 -5
  4. model_library/config/README.md +169 -0
  5. model_library/config/ai21labs_models.yaml +11 -11
  6. model_library/config/alibaba_models.yaml +21 -22
  7. model_library/config/all_models.json +4623 -2599
  8. model_library/config/amazon_models.yaml +100 -102
  9. model_library/config/anthropic_models.yaml +43 -52
  10. model_library/config/cohere_models.yaml +25 -24
  11. model_library/config/deepseek_models.yaml +28 -25
  12. model_library/config/dummy_model.yaml +9 -7
  13. model_library/config/fireworks_models.yaml +86 -56
  14. model_library/config/google_models.yaml +146 -126
  15. model_library/config/inception_models.yaml +6 -6
  16. model_library/config/kimi_models.yaml +13 -14
  17. model_library/config/minimax_models.yaml +37 -0
  18. model_library/config/mistral_models.yaml +85 -29
  19. model_library/config/openai_models.yaml +192 -150
  20. model_library/config/perplexity_models.yaml +10 -23
  21. model_library/config/together_models.yaml +115 -104
  22. model_library/config/xai_models.yaml +47 -79
  23. model_library/config/zai_models.yaml +23 -15
  24. model_library/exceptions.py +7 -16
  25. model_library/providers/amazon.py +32 -17
  26. model_library/providers/minimax.py +33 -0
  27. model_library/providers/mistral.py +10 -1
  28. model_library/providers/openai.py +2 -6
  29. model_library/register_models.py +36 -36
  30. model_library/registry_utils.py +78 -16
  31. model_library/utils.py +2 -2
  32. {model_library-0.1.3.dist-info → model_library-0.1.5.dist-info}/METADATA +2 -2
  33. model_library-0.1.5.dist-info/RECORD +64 -0
  34. model_library-0.1.3.dist-info/RECORD +0 -61
  35. {model_library-0.1.3.dist-info → model_library-0.1.5.dist-info}/WHEEL +0 -0
  36. {model_library-0.1.3.dist-info → model_library-0.1.5.dist-info}/licenses/LICENSE +0 -0
  37. {model_library-0.1.3.dist-info → model_library-0.1.5.dist-info}/top_level.txt +0 -0
model_library/config/together_models.yaml CHANGED
@@ -2,15 +2,17 @@
2
2
  base-config:
3
3
  open_source: true
4
4
  documentation_url: https://docs.together.ai/docs/serverless-models
5
- class_properties:
6
- supports_images: true
5
+ supports:
6
+ images: true
7
+ files: false
8
+ tools: false
9
+ metadata:
7
10
  available_as_evaluator: false
8
- supports_files: false
9
11
  available_for_everyone: true
10
12
  ignored_for_cost: false
11
- supports_tools: false
12
13
  properties:
13
14
  reasoning_model: false
15
+ costs_per_million_token:
14
16
  cache:
15
17
  read_discount: 1
16
18
  batch:
@@ -23,12 +25,11 @@ kimi-models:
23
25
  company: Kimi
24
26
  open_source: true
25
27
  documentation_url: https://www.kimi.com/
26
- class_properties:
27
- supports_tools: true
28
- supports_temperature: true
28
+ supports:
29
+ tools: true
30
+ temperature: true
29
31
  default_parameters:
30
32
  temperature: 0.3
31
- max_output_tokens: 16_384
32
33
 
33
34
  together/moonshotai/Kimi-K2-Instruct:
34
35
  label: Kimi K2 Instruct
@@ -36,11 +37,11 @@ kimi-models:
36
37
  release_date: 2025-07-11
37
38
  properties:
38
39
  context_window: 128_000
39
- max_token_output: 16_384
40
+ max_tokens: 16_384
40
41
  training_cutoff: null
41
42
  reasoning_model: false
42
- class_properties:
43
- supports_images: false
43
+ supports:
44
+ images: false
44
45
  costs_per_million_token:
45
46
  input: 1.00
46
47
  output: 3.00
@@ -50,9 +51,9 @@ llama-4-models:
50
51
  base-config:
51
52
  company: Meta
52
53
  open_source: true
53
- class_properties:
54
- supports_tools: true
55
- supports_images: true
54
+ supports:
55
+ tools: true
56
+ images: true
56
57
 
57
58
  together/meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8:
58
59
  label: Llama 4 Maverick
@@ -61,7 +62,7 @@ llama-4-models:
61
62
  release_date: 2025-04-05
62
63
  properties:
63
64
  context_window: 1_000_000
64
- max_token_output: 16_384
65
+ max_tokens: 16_384
65
66
  training_cutoff: "2024-08"
66
67
  costs_per_million_token:
67
68
  input: 0.27
@@ -75,7 +76,7 @@ llama-4-models:
75
76
  release_date: 2025-04-05
76
77
  properties:
77
78
  context_window: 10_000_000
78
- max_token_output: 16_384
79
+ max_tokens: 16_384
79
80
  training_cutoff: "2024-08"
80
81
  costs_per_million_token:
81
82
  input: 0.18
@@ -84,7 +85,7 @@ llama-4-models:
84
85
  old-llama-models:
85
86
  base-config:
86
87
  company: Meta
87
- class_properties:
88
+ metadata:
88
89
  deprecated: true
89
90
 
90
91
  together/meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo:
@@ -93,7 +94,7 @@ old-llama-models:
93
94
  release_date: 2024-07-23
94
95
  properties:
95
96
  context_window: 131_072
96
- max_token_output: 4_096
97
+ max_tokens: 4_096
97
98
  training_cutoff: "2023-12"
98
99
  costs_per_million_token:
99
100
  input: 0.18
@@ -107,9 +108,9 @@ old-llama-models:
107
108
  release_date: 2024-07-23
108
109
  properties:
109
110
  context_window: 131_072
110
- max_token_output: 4_096
111
+ max_tokens: 4_096
111
112
  training_cutoff: "2023-12"
112
- class_properties:
113
+ metadata:
113
114
  available_as_evaluator: true
114
115
  costs_per_million_token:
115
116
  input: 0.88
@@ -123,7 +124,7 @@ old-llama-models:
123
124
  release_date: 2024-07-23
124
125
  properties:
125
126
  context_window: 130_815
126
- max_token_output: 4_096
127
+ max_tokens: 4_096
127
128
  training_cutoff: "2023-12"
128
129
  costs_per_million_token:
129
130
  input: 3.50
@@ -137,7 +138,7 @@ old-llama-models:
137
138
  release_date: 2024-04-18
138
139
  properties:
139
140
  context_window: 8_192
140
- max_token_output: 4_096
141
+ max_tokens: 4_096
141
142
  training_cutoff: "2023-03"
142
143
  costs_per_million_token:
143
144
  input: 0.20
@@ -151,7 +152,7 @@ old-llama-models:
151
152
  release_date: 2024-04-18
152
153
  properties:
153
154
  context_window: 8_192
154
- max_token_output: 4_096
155
+ max_tokens: 4_096
155
156
  training_cutoff: "2023-12"
156
157
  costs_per_million_token:
157
158
  input: 0.90
@@ -165,7 +166,7 @@ old-llama-models:
165
166
  release_date: 2024-04-18
166
167
  properties:
167
168
  context_window: 8_192
168
- max_token_output: 4_096
169
+ max_tokens: 4_096
169
170
  training_cutoff: "2023-12"
170
171
  costs_per_million_token:
171
172
  input: 0.90
@@ -179,7 +180,7 @@ old-llama-models:
179
180
  release_date: 2024-04-18
180
181
  properties:
181
182
  context_window: 8_192
182
- max_token_output: 4_096
183
+ max_tokens: 4_096
183
184
  training_cutoff: "2023-12"
184
185
  costs_per_million_token:
185
186
  input: 0.90
@@ -193,7 +194,7 @@ old-llama-models:
193
194
  release_date: 2024-04-18
194
195
  properties:
195
196
  context_window: 8_192
196
- max_token_output: 4_096
197
+ max_tokens: 4_096
197
198
  training_cutoff: "2023-12"
198
199
  costs_per_million_token:
199
200
  input: 0.18
@@ -207,7 +208,7 @@ old-llama-models:
207
208
  release_date: 2024-04-18
208
209
  properties:
209
210
  context_window: 8_192
210
- max_token_output: 4_096
211
+ max_tokens: 4_096
211
212
  training_cutoff: "2023-12"
212
213
  costs_per_million_token:
213
214
  input: 0.88
@@ -221,7 +222,7 @@ old-llama-models:
221
222
  release_date: 2024-04-18
222
223
  properties:
223
224
  context_window: 131_072
224
- max_token_output: 4_096
225
+ max_tokens: 4_096
225
226
  training_cutoff: "2023-12"
226
227
  costs_per_million_token:
227
228
  input: 0.06
@@ -235,13 +236,13 @@ old-llama-models:
235
236
  release_date: 2024-04-18
236
237
  properties:
237
238
  context_window: 131_072
238
- max_token_output: 4_096
239
+ max_tokens: 4_096
239
240
  training_cutoff: "2023-12"
240
241
  costs_per_million_token:
241
242
  input: 0.18
242
243
  output: 0.18
243
- class_properties:
244
- supports_images: true
244
+ supports:
245
+ images: true
245
246
  alternative_keys:
246
247
  - together/llama-3.2-11b-instruct
247
248
 
@@ -251,13 +252,14 @@ old-llama-models:
251
252
  release_date: 2024-04-18
252
253
  properties:
253
254
  context_window: 131_072
254
- max_token_output: 4_096
255
+ max_tokens: 4_096
255
256
  training_cutoff: "2023-12"
256
257
  costs_per_million_token:
257
258
  input: 1.2
258
259
  output: 1.2
259
- class_properties:
260
- supports_images: true
260
+ supports:
261
+ images: true
262
+ metadata:
261
263
  deprecated: true
262
264
  alternative_keys:
263
265
  - together/llama-3.2-90b-instruct
@@ -268,10 +270,10 @@ old-llama-models:
268
270
  release_date: 2024-12-06
269
271
  properties:
270
272
  context_window: 128_000
271
- max_token_output: 4_096
273
+ max_tokens: 4_096
272
274
  training_cutoff: "2023-12"
273
- class_properties:
274
- supports_images: false
275
+ supports:
276
+ images: false
275
277
  costs_per_million_token:
276
278
  input: 0.88
277
279
  output: 0.88
@@ -284,7 +286,7 @@ old-llama-models:
284
286
  release_date: 2024-04-18
285
287
  properties:
286
288
  context_window: 8_192
287
- max_token_output: 4_096
289
+ max_tokens: 4_096
288
290
  training_cutoff: "2023-12"
289
291
  costs_per_million_token:
290
292
  input: 0.10
@@ -296,7 +298,7 @@ old-llama-models:
296
298
  release_date: 2024-04-18
297
299
  properties:
298
300
  context_window: 8_192
299
- max_token_output: 4_096
301
+ max_tokens: 4_096
300
302
  training_cutoff: "2023-12"
301
303
  costs_per_million_token:
302
304
  input: 0.54
@@ -308,7 +310,7 @@ old-llama-models:
308
310
  release_date: 2024-04-18
309
311
  properties:
310
312
  context_window: 8_192
311
- max_token_output: 4_096
313
+ max_tokens: 4_096
312
314
  training_cutoff: "2023-12"
313
315
  costs_per_million_token:
314
316
  input: 0.20
@@ -322,7 +324,7 @@ old-llama-models:
322
324
  release_date: 2023-07-18
323
325
  properties:
324
326
  context_window: 4_096
325
- max_token_output: 4_096
327
+ max_tokens: 4_096
326
328
  training_cutoff: "2022-09"
327
329
  costs_per_million_token:
328
330
  input: 0.20
@@ -334,7 +336,7 @@ old-llama-models:
334
336
  release_date: 2023-07-18
335
337
  properties:
336
338
  context_window: 4_096
337
- max_token_output: 4_096
339
+ max_tokens: 4_096
338
340
  training_cutoff: "2022-09"
339
341
  costs_per_million_token:
340
342
  input: 0.20
@@ -348,7 +350,7 @@ old-llama-models:
348
350
  release_date: 2023-07-18
349
351
  properties:
350
352
  context_window: 4_096
351
- max_token_output: 4_096
353
+ max_tokens: 4_096
352
354
  training_cutoff: "2022-09"
353
355
  costs_per_million_token:
354
356
  input: 0.30
@@ -360,7 +362,7 @@ old-llama-models:
360
362
  release_date: 2023-07-18
361
363
  properties:
362
364
  context_window: 4_096
363
- max_token_output: 4_096
365
+ max_tokens: 4_096
364
366
  training_cutoff: "2022-09"
365
367
  costs_per_million_token:
366
368
  input: 0.30
@@ -374,7 +376,7 @@ old-llama-models:
374
376
  release_date: 2023-07-18
375
377
  properties:
376
378
  context_window: 4_096
377
- max_token_output: 4_096
379
+ max_tokens: 4_096
378
380
  training_cutoff: "2022-09"
379
381
  costs_per_million_token:
380
382
  input: 0.90
@@ -386,9 +388,9 @@ old-llama-models:
386
388
  release_date: 2023-07-18
387
389
  properties:
388
390
  context_window: 4_096
389
- max_token_output: 4_096
391
+ max_tokens: 4_096
390
392
  training_cutoff: "2022-09"
391
- class_properties:
393
+ metadata:
392
394
  deprecated: true
393
395
  costs_per_million_token:
394
396
  input: 0.90
@@ -402,7 +404,7 @@ old-llama-models:
402
404
  release_date: 2023-07-18
403
405
  properties:
404
406
  context_window: 4_096
405
- max_token_output: 4_096
407
+ max_tokens: 4_096
406
408
  training_cutoff: "2022-09"
407
409
  costs_per_million_token:
408
410
  input: 0.20
@@ -414,9 +416,9 @@ old-llama-models:
414
416
  release_date: 2023-07-18
415
417
  properties:
416
418
  context_window: 4_096
417
- max_token_output: 4_096
419
+ max_tokens: 4_096
418
420
  training_cutoff: "2022-09"
419
- class_properties:
421
+ metadata:
420
422
  deprecated: true
421
423
  costs_per_million_token:
422
424
  input: 0.20
@@ -430,9 +432,9 @@ old-llama-models:
430
432
  release_date: 2023-07-18
431
433
  properties:
432
434
  context_window: 4_096
433
- max_token_output: 4_096
435
+ max_tokens: 4_096
434
436
  training_cutoff: "2022-09"
435
- class_properties:
437
+ metadata:
436
438
  deprecated: true
437
439
  costs_per_million_token:
438
440
  input: 0.30
@@ -444,7 +446,7 @@ old-llama-models:
444
446
  release_date: 2023-07-18
445
447
  properties:
446
448
  context_window: 4_096
447
- max_token_output: 4_096
449
+ max_tokens: 4_096
448
450
  training_cutoff: "2022-09"
449
451
  costs_per_million_token:
450
452
  input: 0.20
@@ -458,7 +460,7 @@ old-llama-models:
458
460
  release_date: 2023-07-18
459
461
  properties:
460
462
  context_window: 4_096
461
- max_token_output: 4_096
463
+ max_tokens: 4_096
462
464
  training_cutoff: "2022-09"
463
465
  costs_per_million_token:
464
466
  input: 0.90
@@ -470,7 +472,7 @@ old-llama-models:
470
472
  nvidia-models:
471
473
  base-config:
472
474
  company: NVIDIA
473
- class_properties:
475
+ metadata:
474
476
  deprecated: true
475
477
 
476
478
  together/nvidia/Llama-3.1-Nemotron-70B-Instruct-HF:
@@ -479,7 +481,7 @@ nvidia-models:
479
481
  release_date: 2024-04-18
480
482
  properties:
481
483
  context_window: 32_768
482
- max_token_output: 4_096
484
+ max_tokens: 4_096
483
485
  training_cutoff: "2023-12"
484
486
  costs_per_million_token:
485
487
  input: 0.90
@@ -488,9 +490,10 @@ nvidia-models:
488
490
  qwen-models:
489
491
  base-config:
490
492
  company: Alibaba
491
- class_properties:
493
+ supports:
494
+ temperature: true
495
+ metadata:
492
496
  deprecated: true
493
- supports_temperature: true
494
497
  default_parameters:
495
498
  temperature: 0.7
496
499
 
@@ -500,14 +503,15 @@ qwen-models:
500
503
  release_date: 2024-04-18
501
504
  properties:
502
505
  context_window: 32_768
503
- max_token_output: 4_096
506
+ max_tokens: 4_096
504
507
  training_cutoff: "2023-12"
505
508
  costs_per_million_token:
506
509
  input: 0.80
507
510
  output: 0.80
508
- class_properties:
511
+ supports:
512
+ images: false
513
+ metadata:
509
514
  available_for_everyone: false
510
- supports_images: false
511
515
 
512
516
  together/Qwen/Qwen2.5-7B-Instruct-Turbo:
513
517
  label: Qwen 2.5 Instruct Turbo (7B)
@@ -515,14 +519,15 @@ qwen-models:
515
519
  release_date: 2024-04-18
516
520
  properties:
517
521
  context_window: 32_768
518
- max_token_output: 4_096
522
+ max_tokens: 4_096
519
523
  training_cutoff: "2023-12"
520
524
  costs_per_million_token:
521
525
  input: 0.30
522
526
  output: 0.30
523
- class_properties:
527
+ supports:
528
+ images: false
529
+ metadata:
524
530
  available_for_everyone: false
525
- supports_images: false
526
531
  alternative_keys:
527
532
  - together/qwen-2.5-7b-instruct
528
533
 
@@ -532,14 +537,15 @@ qwen-models:
532
537
  release_date: 2024-04-18
533
538
  properties:
534
539
  context_window: 32_768
535
- max_token_output: 4_096
540
+ max_tokens: 4_096
536
541
  training_cutoff: "2023-12"
537
542
  costs_per_million_token:
538
543
  input: 1.20
539
544
  output: 1.20
540
- class_properties:
545
+ supports:
546
+ images: false
547
+ metadata:
541
548
  available_for_everyone: false
542
- supports_images: false
543
549
  alternative_keys:
544
550
  - together/qwen-2.5-72b-instruct
545
551
 
@@ -547,9 +553,10 @@ qwen-models:
547
553
  mistralai-models:
548
554
  base-config:
549
555
  company: Mistral
550
- class_properties:
556
+ supports:
557
+ images: false
558
+ metadata:
551
559
  deprecated: true
552
- supports_images: false
553
560
 
554
561
  together/mistralai/Mistral-7B-v0.1:
555
562
  label: Mistral (7B)
@@ -557,7 +564,7 @@ mistralai-models:
557
564
  release_date: 2023-09-27
558
565
  properties:
559
566
  context_window: 8_192
560
- max_token_output: 4_096
567
+ max_tokens: 4_096
561
568
  training_cutoff: "2023-06"
562
569
  costs_per_million_token:
563
570
  input: 0.18
@@ -571,7 +578,7 @@ mistralai-models:
571
578
  release_date: 2023-09-27
572
579
  properties:
573
580
  context_window: 32_768
574
- max_token_output: 4_096
581
+ max_tokens: 4_096
575
582
  training_cutoff: "2023-06"
576
583
  costs_per_million_token:
577
584
  input: 0.18
@@ -585,11 +592,11 @@ mistralai-models:
585
592
  release_date: 2023-12-15
586
593
  properties:
587
594
  context_window: 32_768
588
- max_token_output: 4_096
595
+ max_tokens: 4_096
589
596
  training_cutoff: "2023-09"
590
597
  costs_per_million_token:
591
- input: 0.18
592
- output: 0.18
598
+ input: 0.2
599
+ output: 0.2
593
600
  alternative_keys:
594
601
  - together/Mistral-7B-Instruct-v0.2
595
602
 
@@ -599,7 +606,7 @@ mistralai-models:
599
606
  release_date: 2024-04-18
600
607
  properties:
601
608
  context_window: 32_768
602
- max_token_output: 4_096
609
+ max_tokens: 4_096
603
610
  training_cutoff: "2023-12"
604
611
  costs_per_million_token:
605
612
  input: 0.18
@@ -613,7 +620,7 @@ mistralai-models:
613
620
  release_date: 2023-12-15
614
621
  properties:
615
622
  context_window: 32_768
616
- max_token_output: 4_096
623
+ max_tokens: 4_096
617
624
  training_cutoff: "2023-09"
618
625
  costs_per_million_token:
619
626
  input: 0.60
@@ -627,7 +634,7 @@ mistralai-models:
627
634
  release_date: 2023-12-15
628
635
  properties:
629
636
  context_window: 32_768
630
- max_token_output: 4_096
637
+ max_tokens: 4_096
631
638
  training_cutoff: "2023-09"
632
639
  costs_per_million_token:
633
640
  input: 0.60
@@ -641,9 +648,9 @@ mistralai-models:
641
648
  release_date: 2024-02-15
642
649
  properties:
643
650
  context_window: 65_536
644
- max_token_output: 4_096
651
+ max_tokens: 4_096
645
652
  training_cutoff: "2023-12"
646
- class_properties:
653
+ metadata:
647
654
  available_as_evaluator: true
648
655
  costs_per_million_token:
649
656
  input: 1.20
@@ -661,10 +668,11 @@ meta-llama-vision:
661
668
  release_date: 2024-04-18
662
669
  properties:
663
670
  context_window: 131_072
664
- max_token_output: 4_096
671
+ max_tokens: 4_096
665
672
  training_cutoff: "2023-12"
666
- class_properties:
667
- supports_images: true
673
+ supports:
674
+ images: true
675
+ metadata:
668
676
  available_for_everyone: false
669
677
  deprecated: true
670
678
 
@@ -675,10 +683,11 @@ meta-llama-vision:
675
683
  release_date: 2024-04-18
676
684
  properties:
677
685
  context_window: 131_072
678
- max_token_output: 4_096
686
+ max_tokens: 4_096
679
687
  training_cutoff: "2023-12"
680
- class_properties:
681
- supports_images: true
688
+ supports:
689
+ images: true
690
+ metadata:
682
691
  available_for_everyone: true
683
692
  deprecated: true
684
693
  costs_per_million_token:
@@ -694,10 +703,11 @@ meta-llama-vision:
694
703
  release_date: 2024-04-18
695
704
  properties:
696
705
  context_window: 131_072
697
- max_token_output: 4_096
706
+ max_tokens: 4_096
698
707
  training_cutoff: "2023-12"
699
- class_properties:
700
- supports_images: true
708
+ supports:
709
+ images: true
710
+ metadata:
701
711
  available_for_everyone: true
702
712
  deprecated: true
703
713
  costs_per_million_token:
@@ -715,15 +725,16 @@ google-models:
715
725
  release_date: 2024-02-21
716
726
  properties:
717
727
  context_window: 8_192
718
- max_token_output: 4_096
728
+ max_tokens: 4_096
719
729
  training_cutoff: "2023-12"
720
730
  costs_per_million_token:
721
731
  input: 0.50
722
732
  output: 0.50
723
- class_properties:
733
+ metadata:
724
734
  available_for_everyone: false
725
735
  deprecated: true
726
- supports_images: false
736
+ supports:
737
+ images: false
727
738
  alternative_keys:
728
739
  - together/gemma-2-27b-instruct
729
740
 
@@ -734,12 +745,12 @@ google-models:
734
745
  release_date: 2024-02-21
735
746
  properties:
736
747
  context_window: 8_192
737
- max_token_output: 4_096
748
+ max_tokens: 4_096
738
749
  training_cutoff: "2023-12"
739
750
  costs_per_million_token:
740
751
  input: 0.20
741
752
  output: 0.20
742
- class_properties:
753
+ metadata:
743
754
  available_for_everyone: false
744
755
  deprecated: true
745
756
  alternative_keys:
@@ -752,12 +763,12 @@ google-models:
752
763
  release_date: 2024-02-21
753
764
  properties:
754
765
  context_window: 8_192
755
- max_token_output: 4_096
766
+ max_tokens: 4_096
756
767
  training_cutoff: "2023-12"
757
768
  costs_per_million_token:
758
769
  input: 0.10
759
770
  output: 0.10
760
- class_properties:
771
+ metadata:
761
772
  available_for_everyone: false
762
773
  deprecated: true
763
774
  alternative_keys:
@@ -767,7 +778,7 @@ google-models:
767
778
  falcon-models:
768
779
  base-config:
769
780
  company: Technology Innovation Institute
770
- class_properties:
781
+ metadata:
771
782
  deprecated: true
772
783
 
773
784
  together/togethercomputer/falcon-7b-instruct:
@@ -820,7 +831,7 @@ alpaca-models:
820
831
  costs_per_million_token:
821
832
  input: 0.20
822
833
  output: 0.20
823
- class_properties:
834
+ metadata:
824
835
  deprecated: true
825
836
  alternative_keys:
826
837
  - together/alpaca-7b
@@ -830,9 +841,9 @@ deepseek-models:
830
841
  base-config:
831
842
  company: DeepSeek
832
843
  open_source: true
833
- class_properties:
834
- supports_images: false
835
- supports_temperature: true
844
+ supports:
845
+ images: false
846
+ temperature: true
836
847
  default_parameters:
837
848
  temperature: 1
838
849
 
@@ -842,9 +853,9 @@ deepseek-models:
842
853
  release_date: 2024-12-26
843
854
  properties:
844
855
  context_window: 131_072
845
- max_token_output: null
856
+ max_tokens: null
846
857
  training_cutoff: null
847
- class_properties:
858
+ metadata:
848
859
  deprecated: true
849
860
  costs_per_million_token:
850
861
  input: 1.25
@@ -857,10 +868,10 @@ deepseek-models:
857
868
  release_date: 2025-01-20
858
869
  properties:
859
870
  context_window: 163_840
860
- max_token_output: null
871
+ max_tokens: null
861
872
  training_cutoff: null
862
873
  reasoning_model: true
863
- class_properties:
874
+ metadata:
864
875
  deprecated: true
865
876
  costs_per_million_token:
866
877
  input: 3.00