model-library 0.1.2__py3-none-any.whl → 0.1.4__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 (41) hide show
  1. model_library/base/base.py +13 -6
  2. model_library/base/output.py +55 -0
  3. model_library/base/utils.py +3 -2
  4. model_library/config/README.md +169 -0
  5. model_library/config/ai21labs_models.yaml +11 -10
  6. model_library/config/alibaba_models.yaml +21 -22
  7. model_library/config/all_models.json +4708 -2471
  8. model_library/config/amazon_models.yaml +100 -102
  9. model_library/config/anthropic_models.yaml +59 -45
  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 +156 -102
  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 -159
  20. model_library/config/perplexity_models.yaml +8 -23
  21. model_library/config/together_models.yaml +115 -103
  22. model_library/config/xai_models.yaml +85 -57
  23. model_library/config/zai_models.yaml +23 -15
  24. model_library/exceptions.py +12 -17
  25. model_library/file_utils.py +1 -1
  26. model_library/providers/amazon.py +32 -17
  27. model_library/providers/anthropic.py +2 -6
  28. model_library/providers/google/google.py +35 -29
  29. model_library/providers/minimax.py +33 -0
  30. model_library/providers/mistral.py +10 -1
  31. model_library/providers/openai.py +10 -8
  32. model_library/providers/together.py +18 -211
  33. model_library/register_models.py +36 -38
  34. model_library/registry_utils.py +18 -16
  35. model_library/utils.py +2 -2
  36. {model_library-0.1.2.dist-info → model_library-0.1.4.dist-info}/METADATA +3 -4
  37. model_library-0.1.4.dist-info/RECORD +64 -0
  38. model_library-0.1.2.dist-info/RECORD +0 -61
  39. {model_library-0.1.2.dist-info → model_library-0.1.4.dist-info}/WHEEL +0 -0
  40. {model_library-0.1.2.dist-info → model_library-0.1.4.dist-info}/licenses/LICENSE +0 -0
  41. {model_library-0.1.2.dist-info → model_library-0.1.4.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:
@@ -23,9 +25,9 @@ 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
33
 
@@ -35,11 +37,11 @@ kimi-models:
35
37
  release_date: 2025-07-11
36
38
  properties:
37
39
  context_window: 128_000
38
- max_token_output: 16_384
40
+ max_tokens: 16_384
39
41
  training_cutoff: null
40
42
  reasoning_model: false
41
- class_properties:
42
- supports_images: false
43
+ supports:
44
+ images: false
43
45
  costs_per_million_token:
44
46
  input: 1.00
45
47
  output: 3.00
@@ -49,9 +51,9 @@ llama-4-models:
49
51
  base-config:
50
52
  company: Meta
51
53
  open_source: true
52
- class_properties:
53
- supports_tools: true
54
- supports_images: true
54
+ supports:
55
+ tools: true
56
+ images: true
55
57
 
56
58
  together/meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8:
57
59
  label: Llama 4 Maverick
@@ -60,7 +62,7 @@ llama-4-models:
60
62
  release_date: 2025-04-05
61
63
  properties:
62
64
  context_window: 1_000_000
63
- max_token_output: 16_384
65
+ max_tokens: 16_384
64
66
  training_cutoff: "2024-08"
65
67
  costs_per_million_token:
66
68
  input: 0.27
@@ -74,7 +76,7 @@ llama-4-models:
74
76
  release_date: 2025-04-05
75
77
  properties:
76
78
  context_window: 10_000_000
77
- max_token_output: 16_384
79
+ max_tokens: 16_384
78
80
  training_cutoff: "2024-08"
79
81
  costs_per_million_token:
80
82
  input: 0.18
@@ -83,7 +85,7 @@ llama-4-models:
83
85
  old-llama-models:
84
86
  base-config:
85
87
  company: Meta
86
- class_properties:
88
+ metadata:
87
89
  deprecated: true
88
90
 
89
91
  together/meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo:
@@ -92,7 +94,7 @@ old-llama-models:
92
94
  release_date: 2024-07-23
93
95
  properties:
94
96
  context_window: 131_072
95
- max_token_output: 4_096
97
+ max_tokens: 4_096
96
98
  training_cutoff: "2023-12"
97
99
  costs_per_million_token:
98
100
  input: 0.18
@@ -106,9 +108,9 @@ old-llama-models:
106
108
  release_date: 2024-07-23
107
109
  properties:
108
110
  context_window: 131_072
109
- max_token_output: 4_096
111
+ max_tokens: 4_096
110
112
  training_cutoff: "2023-12"
111
- class_properties:
113
+ metadata:
112
114
  available_as_evaluator: true
113
115
  costs_per_million_token:
114
116
  input: 0.88
@@ -122,7 +124,7 @@ old-llama-models:
122
124
  release_date: 2024-07-23
123
125
  properties:
124
126
  context_window: 130_815
125
- max_token_output: 4_096
127
+ max_tokens: 4_096
126
128
  training_cutoff: "2023-12"
127
129
  costs_per_million_token:
128
130
  input: 3.50
@@ -136,7 +138,7 @@ old-llama-models:
136
138
  release_date: 2024-04-18
137
139
  properties:
138
140
  context_window: 8_192
139
- max_token_output: 4_096
141
+ max_tokens: 4_096
140
142
  training_cutoff: "2023-03"
141
143
  costs_per_million_token:
142
144
  input: 0.20
@@ -150,7 +152,7 @@ old-llama-models:
150
152
  release_date: 2024-04-18
151
153
  properties:
152
154
  context_window: 8_192
153
- max_token_output: 4_096
155
+ max_tokens: 4_096
154
156
  training_cutoff: "2023-12"
155
157
  costs_per_million_token:
156
158
  input: 0.90
@@ -164,7 +166,7 @@ old-llama-models:
164
166
  release_date: 2024-04-18
165
167
  properties:
166
168
  context_window: 8_192
167
- max_token_output: 4_096
169
+ max_tokens: 4_096
168
170
  training_cutoff: "2023-12"
169
171
  costs_per_million_token:
170
172
  input: 0.90
@@ -178,7 +180,7 @@ old-llama-models:
178
180
  release_date: 2024-04-18
179
181
  properties:
180
182
  context_window: 8_192
181
- max_token_output: 4_096
183
+ max_tokens: 4_096
182
184
  training_cutoff: "2023-12"
183
185
  costs_per_million_token:
184
186
  input: 0.90
@@ -192,7 +194,7 @@ old-llama-models:
192
194
  release_date: 2024-04-18
193
195
  properties:
194
196
  context_window: 8_192
195
- max_token_output: 4_096
197
+ max_tokens: 4_096
196
198
  training_cutoff: "2023-12"
197
199
  costs_per_million_token:
198
200
  input: 0.18
@@ -206,7 +208,7 @@ old-llama-models:
206
208
  release_date: 2024-04-18
207
209
  properties:
208
210
  context_window: 8_192
209
- max_token_output: 4_096
211
+ max_tokens: 4_096
210
212
  training_cutoff: "2023-12"
211
213
  costs_per_million_token:
212
214
  input: 0.88
@@ -220,7 +222,7 @@ old-llama-models:
220
222
  release_date: 2024-04-18
221
223
  properties:
222
224
  context_window: 131_072
223
- max_token_output: 4_096
225
+ max_tokens: 4_096
224
226
  training_cutoff: "2023-12"
225
227
  costs_per_million_token:
226
228
  input: 0.06
@@ -234,13 +236,13 @@ old-llama-models:
234
236
  release_date: 2024-04-18
235
237
  properties:
236
238
  context_window: 131_072
237
- max_token_output: 4_096
239
+ max_tokens: 4_096
238
240
  training_cutoff: "2023-12"
239
241
  costs_per_million_token:
240
242
  input: 0.18
241
243
  output: 0.18
242
- class_properties:
243
- supports_images: true
244
+ supports:
245
+ images: true
244
246
  alternative_keys:
245
247
  - together/llama-3.2-11b-instruct
246
248
 
@@ -250,13 +252,14 @@ old-llama-models:
250
252
  release_date: 2024-04-18
251
253
  properties:
252
254
  context_window: 131_072
253
- max_token_output: 4_096
255
+ max_tokens: 4_096
254
256
  training_cutoff: "2023-12"
255
257
  costs_per_million_token:
256
258
  input: 1.2
257
259
  output: 1.2
258
- class_properties:
259
- supports_images: true
260
+ supports:
261
+ images: true
262
+ metadata:
260
263
  deprecated: true
261
264
  alternative_keys:
262
265
  - together/llama-3.2-90b-instruct
@@ -267,10 +270,10 @@ old-llama-models:
267
270
  release_date: 2024-12-06
268
271
  properties:
269
272
  context_window: 128_000
270
- max_token_output: 4_096
273
+ max_tokens: 4_096
271
274
  training_cutoff: "2023-12"
272
- class_properties:
273
- supports_images: false
275
+ supports:
276
+ images: false
274
277
  costs_per_million_token:
275
278
  input: 0.88
276
279
  output: 0.88
@@ -283,7 +286,7 @@ old-llama-models:
283
286
  release_date: 2024-04-18
284
287
  properties:
285
288
  context_window: 8_192
286
- max_token_output: 4_096
289
+ max_tokens: 4_096
287
290
  training_cutoff: "2023-12"
288
291
  costs_per_million_token:
289
292
  input: 0.10
@@ -295,7 +298,7 @@ old-llama-models:
295
298
  release_date: 2024-04-18
296
299
  properties:
297
300
  context_window: 8_192
298
- max_token_output: 4_096
301
+ max_tokens: 4_096
299
302
  training_cutoff: "2023-12"
300
303
  costs_per_million_token:
301
304
  input: 0.54
@@ -307,7 +310,7 @@ old-llama-models:
307
310
  release_date: 2024-04-18
308
311
  properties:
309
312
  context_window: 8_192
310
- max_token_output: 4_096
313
+ max_tokens: 4_096
311
314
  training_cutoff: "2023-12"
312
315
  costs_per_million_token:
313
316
  input: 0.20
@@ -321,7 +324,7 @@ old-llama-models:
321
324
  release_date: 2023-07-18
322
325
  properties:
323
326
  context_window: 4_096
324
- max_token_output: 4_096
327
+ max_tokens: 4_096
325
328
  training_cutoff: "2022-09"
326
329
  costs_per_million_token:
327
330
  input: 0.20
@@ -333,7 +336,7 @@ old-llama-models:
333
336
  release_date: 2023-07-18
334
337
  properties:
335
338
  context_window: 4_096
336
- max_token_output: 4_096
339
+ max_tokens: 4_096
337
340
  training_cutoff: "2022-09"
338
341
  costs_per_million_token:
339
342
  input: 0.20
@@ -347,7 +350,7 @@ old-llama-models:
347
350
  release_date: 2023-07-18
348
351
  properties:
349
352
  context_window: 4_096
350
- max_token_output: 4_096
353
+ max_tokens: 4_096
351
354
  training_cutoff: "2022-09"
352
355
  costs_per_million_token:
353
356
  input: 0.30
@@ -359,7 +362,7 @@ old-llama-models:
359
362
  release_date: 2023-07-18
360
363
  properties:
361
364
  context_window: 4_096
362
- max_token_output: 4_096
365
+ max_tokens: 4_096
363
366
  training_cutoff: "2022-09"
364
367
  costs_per_million_token:
365
368
  input: 0.30
@@ -373,7 +376,7 @@ old-llama-models:
373
376
  release_date: 2023-07-18
374
377
  properties:
375
378
  context_window: 4_096
376
- max_token_output: 4_096
379
+ max_tokens: 4_096
377
380
  training_cutoff: "2022-09"
378
381
  costs_per_million_token:
379
382
  input: 0.90
@@ -385,9 +388,9 @@ old-llama-models:
385
388
  release_date: 2023-07-18
386
389
  properties:
387
390
  context_window: 4_096
388
- max_token_output: 4_096
391
+ max_tokens: 4_096
389
392
  training_cutoff: "2022-09"
390
- class_properties:
393
+ metadata:
391
394
  deprecated: true
392
395
  costs_per_million_token:
393
396
  input: 0.90
@@ -401,7 +404,7 @@ old-llama-models:
401
404
  release_date: 2023-07-18
402
405
  properties:
403
406
  context_window: 4_096
404
- max_token_output: 4_096
407
+ max_tokens: 4_096
405
408
  training_cutoff: "2022-09"
406
409
  costs_per_million_token:
407
410
  input: 0.20
@@ -413,9 +416,9 @@ old-llama-models:
413
416
  release_date: 2023-07-18
414
417
  properties:
415
418
  context_window: 4_096
416
- max_token_output: 4_096
419
+ max_tokens: 4_096
417
420
  training_cutoff: "2022-09"
418
- class_properties:
421
+ metadata:
419
422
  deprecated: true
420
423
  costs_per_million_token:
421
424
  input: 0.20
@@ -429,9 +432,9 @@ old-llama-models:
429
432
  release_date: 2023-07-18
430
433
  properties:
431
434
  context_window: 4_096
432
- max_token_output: 4_096
435
+ max_tokens: 4_096
433
436
  training_cutoff: "2022-09"
434
- class_properties:
437
+ metadata:
435
438
  deprecated: true
436
439
  costs_per_million_token:
437
440
  input: 0.30
@@ -443,7 +446,7 @@ old-llama-models:
443
446
  release_date: 2023-07-18
444
447
  properties:
445
448
  context_window: 4_096
446
- max_token_output: 4_096
449
+ max_tokens: 4_096
447
450
  training_cutoff: "2022-09"
448
451
  costs_per_million_token:
449
452
  input: 0.20
@@ -457,7 +460,7 @@ old-llama-models:
457
460
  release_date: 2023-07-18
458
461
  properties:
459
462
  context_window: 4_096
460
- max_token_output: 4_096
463
+ max_tokens: 4_096
461
464
  training_cutoff: "2022-09"
462
465
  costs_per_million_token:
463
466
  input: 0.90
@@ -469,7 +472,7 @@ old-llama-models:
469
472
  nvidia-models:
470
473
  base-config:
471
474
  company: NVIDIA
472
- class_properties:
475
+ metadata:
473
476
  deprecated: true
474
477
 
475
478
  together/nvidia/Llama-3.1-Nemotron-70B-Instruct-HF:
@@ -478,7 +481,7 @@ nvidia-models:
478
481
  release_date: 2024-04-18
479
482
  properties:
480
483
  context_window: 32_768
481
- max_token_output: 4_096
484
+ max_tokens: 4_096
482
485
  training_cutoff: "2023-12"
483
486
  costs_per_million_token:
484
487
  input: 0.90
@@ -487,9 +490,10 @@ nvidia-models:
487
490
  qwen-models:
488
491
  base-config:
489
492
  company: Alibaba
490
- class_properties:
493
+ supports:
494
+ temperature: true
495
+ metadata:
491
496
  deprecated: true
492
- supports_temperature: true
493
497
  default_parameters:
494
498
  temperature: 0.7
495
499
 
@@ -499,14 +503,15 @@ qwen-models:
499
503
  release_date: 2024-04-18
500
504
  properties:
501
505
  context_window: 32_768
502
- max_token_output: 4_096
506
+ max_tokens: 4_096
503
507
  training_cutoff: "2023-12"
504
508
  costs_per_million_token:
505
509
  input: 0.80
506
510
  output: 0.80
507
- class_properties:
511
+ supports:
512
+ images: false
513
+ metadata:
508
514
  available_for_everyone: false
509
- supports_images: false
510
515
 
511
516
  together/Qwen/Qwen2.5-7B-Instruct-Turbo:
512
517
  label: Qwen 2.5 Instruct Turbo (7B)
@@ -514,14 +519,15 @@ qwen-models:
514
519
  release_date: 2024-04-18
515
520
  properties:
516
521
  context_window: 32_768
517
- max_token_output: 4_096
522
+ max_tokens: 4_096
518
523
  training_cutoff: "2023-12"
519
524
  costs_per_million_token:
520
525
  input: 0.30
521
526
  output: 0.30
522
- class_properties:
527
+ supports:
528
+ images: false
529
+ metadata:
523
530
  available_for_everyone: false
524
- supports_images: false
525
531
  alternative_keys:
526
532
  - together/qwen-2.5-7b-instruct
527
533
 
@@ -531,14 +537,15 @@ qwen-models:
531
537
  release_date: 2024-04-18
532
538
  properties:
533
539
  context_window: 32_768
534
- max_token_output: 4_096
540
+ max_tokens: 4_096
535
541
  training_cutoff: "2023-12"
536
542
  costs_per_million_token:
537
543
  input: 1.20
538
544
  output: 1.20
539
- class_properties:
545
+ supports:
546
+ images: false
547
+ metadata:
540
548
  available_for_everyone: false
541
- supports_images: false
542
549
  alternative_keys:
543
550
  - together/qwen-2.5-72b-instruct
544
551
 
@@ -546,9 +553,10 @@ qwen-models:
546
553
  mistralai-models:
547
554
  base-config:
548
555
  company: Mistral
549
- class_properties:
556
+ supports:
557
+ images: false
558
+ metadata:
550
559
  deprecated: true
551
- supports_images: false
552
560
 
553
561
  together/mistralai/Mistral-7B-v0.1:
554
562
  label: Mistral (7B)
@@ -556,7 +564,7 @@ mistralai-models:
556
564
  release_date: 2023-09-27
557
565
  properties:
558
566
  context_window: 8_192
559
- max_token_output: 4_096
567
+ max_tokens: 4_096
560
568
  training_cutoff: "2023-06"
561
569
  costs_per_million_token:
562
570
  input: 0.18
@@ -570,7 +578,7 @@ mistralai-models:
570
578
  release_date: 2023-09-27
571
579
  properties:
572
580
  context_window: 32_768
573
- max_token_output: 4_096
581
+ max_tokens: 4_096
574
582
  training_cutoff: "2023-06"
575
583
  costs_per_million_token:
576
584
  input: 0.18
@@ -584,11 +592,11 @@ mistralai-models:
584
592
  release_date: 2023-12-15
585
593
  properties:
586
594
  context_window: 32_768
587
- max_token_output: 4_096
595
+ max_tokens: 4_096
588
596
  training_cutoff: "2023-09"
589
597
  costs_per_million_token:
590
- input: 0.18
591
- output: 0.18
598
+ input: 0.2
599
+ output: 0.2
592
600
  alternative_keys:
593
601
  - together/Mistral-7B-Instruct-v0.2
594
602
 
@@ -598,7 +606,7 @@ mistralai-models:
598
606
  release_date: 2024-04-18
599
607
  properties:
600
608
  context_window: 32_768
601
- max_token_output: 4_096
609
+ max_tokens: 4_096
602
610
  training_cutoff: "2023-12"
603
611
  costs_per_million_token:
604
612
  input: 0.18
@@ -612,7 +620,7 @@ mistralai-models:
612
620
  release_date: 2023-12-15
613
621
  properties:
614
622
  context_window: 32_768
615
- max_token_output: 4_096
623
+ max_tokens: 4_096
616
624
  training_cutoff: "2023-09"
617
625
  costs_per_million_token:
618
626
  input: 0.60
@@ -626,7 +634,7 @@ mistralai-models:
626
634
  release_date: 2023-12-15
627
635
  properties:
628
636
  context_window: 32_768
629
- max_token_output: 4_096
637
+ max_tokens: 4_096
630
638
  training_cutoff: "2023-09"
631
639
  costs_per_million_token:
632
640
  input: 0.60
@@ -640,9 +648,9 @@ mistralai-models:
640
648
  release_date: 2024-02-15
641
649
  properties:
642
650
  context_window: 65_536
643
- max_token_output: 4_096
651
+ max_tokens: 4_096
644
652
  training_cutoff: "2023-12"
645
- class_properties:
653
+ metadata:
646
654
  available_as_evaluator: true
647
655
  costs_per_million_token:
648
656
  input: 1.20
@@ -660,10 +668,11 @@ meta-llama-vision:
660
668
  release_date: 2024-04-18
661
669
  properties:
662
670
  context_window: 131_072
663
- max_token_output: 4_096
671
+ max_tokens: 4_096
664
672
  training_cutoff: "2023-12"
665
- class_properties:
666
- supports_images: true
673
+ supports:
674
+ images: true
675
+ metadata:
667
676
  available_for_everyone: false
668
677
  deprecated: true
669
678
 
@@ -674,10 +683,11 @@ meta-llama-vision:
674
683
  release_date: 2024-04-18
675
684
  properties:
676
685
  context_window: 131_072
677
- max_token_output: 4_096
686
+ max_tokens: 4_096
678
687
  training_cutoff: "2023-12"
679
- class_properties:
680
- supports_images: true
688
+ supports:
689
+ images: true
690
+ metadata:
681
691
  available_for_everyone: true
682
692
  deprecated: true
683
693
  costs_per_million_token:
@@ -693,10 +703,11 @@ meta-llama-vision:
693
703
  release_date: 2024-04-18
694
704
  properties:
695
705
  context_window: 131_072
696
- max_token_output: 4_096
706
+ max_tokens: 4_096
697
707
  training_cutoff: "2023-12"
698
- class_properties:
699
- supports_images: true
708
+ supports:
709
+ images: true
710
+ metadata:
700
711
  available_for_everyone: true
701
712
  deprecated: true
702
713
  costs_per_million_token:
@@ -714,15 +725,16 @@ google-models:
714
725
  release_date: 2024-02-21
715
726
  properties:
716
727
  context_window: 8_192
717
- max_token_output: 4_096
728
+ max_tokens: 4_096
718
729
  training_cutoff: "2023-12"
719
730
  costs_per_million_token:
720
731
  input: 0.50
721
732
  output: 0.50
722
- class_properties:
733
+ metadata:
723
734
  available_for_everyone: false
724
735
  deprecated: true
725
- supports_images: false
736
+ supports:
737
+ images: false
726
738
  alternative_keys:
727
739
  - together/gemma-2-27b-instruct
728
740
 
@@ -733,12 +745,12 @@ google-models:
733
745
  release_date: 2024-02-21
734
746
  properties:
735
747
  context_window: 8_192
736
- max_token_output: 4_096
748
+ max_tokens: 4_096
737
749
  training_cutoff: "2023-12"
738
750
  costs_per_million_token:
739
751
  input: 0.20
740
752
  output: 0.20
741
- class_properties:
753
+ metadata:
742
754
  available_for_everyone: false
743
755
  deprecated: true
744
756
  alternative_keys:
@@ -751,12 +763,12 @@ google-models:
751
763
  release_date: 2024-02-21
752
764
  properties:
753
765
  context_window: 8_192
754
- max_token_output: 4_096
766
+ max_tokens: 4_096
755
767
  training_cutoff: "2023-12"
756
768
  costs_per_million_token:
757
769
  input: 0.10
758
770
  output: 0.10
759
- class_properties:
771
+ metadata:
760
772
  available_for_everyone: false
761
773
  deprecated: true
762
774
  alternative_keys:
@@ -766,7 +778,7 @@ google-models:
766
778
  falcon-models:
767
779
  base-config:
768
780
  company: Technology Innovation Institute
769
- class_properties:
781
+ metadata:
770
782
  deprecated: true
771
783
 
772
784
  together/togethercomputer/falcon-7b-instruct:
@@ -819,7 +831,7 @@ alpaca-models:
819
831
  costs_per_million_token:
820
832
  input: 0.20
821
833
  output: 0.20
822
- class_properties:
834
+ metadata:
823
835
  deprecated: true
824
836
  alternative_keys:
825
837
  - together/alpaca-7b
@@ -829,9 +841,9 @@ deepseek-models:
829
841
  base-config:
830
842
  company: DeepSeek
831
843
  open_source: true
832
- class_properties:
833
- supports_images: false
834
- supports_temperature: true
844
+ supports:
845
+ images: false
846
+ temperature: true
835
847
  default_parameters:
836
848
  temperature: 1
837
849
 
@@ -841,9 +853,9 @@ deepseek-models:
841
853
  release_date: 2024-12-26
842
854
  properties:
843
855
  context_window: 131_072
844
- max_token_output: null
856
+ max_tokens: null
845
857
  training_cutoff: null
846
- class_properties:
858
+ metadata:
847
859
  deprecated: true
848
860
  costs_per_million_token:
849
861
  input: 1.25
@@ -856,10 +868,10 @@ deepseek-models:
856
868
  release_date: 2025-01-20
857
869
  properties:
858
870
  context_window: 163_840
859
- max_token_output: null
871
+ max_tokens: null
860
872
  training_cutoff: null
861
873
  reasoning_model: true
862
- class_properties:
874
+ metadata:
863
875
  deprecated: true
864
876
  costs_per_million_token:
865
877
  input: 3.00