@lobehub/lobehub 2.0.0-next.170 → 2.0.0-next.172
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/CHANGELOG.md +50 -0
- package/README.md +8 -8
- package/README.zh-CN.md +8 -8
- package/changelog/v1.json +18 -0
- package/docs/development/database-schema.dbml +33 -0
- package/locales/ar/models.json +12 -0
- package/locales/bg-BG/models.json +12 -0
- package/locales/de-DE/models.json +12 -0
- package/locales/en-US/models.json +12 -0
- package/locales/es-ES/models.json +12 -0
- package/locales/fa-IR/models.json +12 -0
- package/locales/fr-FR/models.json +12 -0
- package/locales/it-IT/models.json +12 -0
- package/locales/ja-JP/models.json +12 -0
- package/locales/ko-KR/models.json +12 -0
- package/locales/nl-NL/models.json +12 -0
- package/locales/pl-PL/models.json +12 -0
- package/locales/pt-BR/models.json +12 -0
- package/locales/ru-RU/models.json +12 -0
- package/locales/tr-TR/models.json +12 -0
- package/locales/vi-VN/models.json +12 -0
- package/locales/zh-CN/models.json +12 -0
- package/locales/zh-TW/models.json +12 -0
- package/package.json +1 -1
- package/packages/database/migrations/0062_add_more_index.sql +15 -0
- package/packages/database/migrations/meta/0062_snapshot.json +8960 -0
- package/packages/database/migrations/meta/_journal.json +7 -0
- package/packages/database/src/core/migrations.json +42 -20
- package/packages/database/src/schemas/apiKey.ts +17 -13
- package/packages/database/src/schemas/asyncTask.ts +16 -12
- package/packages/database/src/schemas/chatGroup.ts +5 -1
- package/packages/database/src/schemas/file.ts +24 -23
- package/packages/database/src/schemas/generation.ts +90 -72
- package/packages/database/src/schemas/message.ts +1 -0
- package/packages/database/src/schemas/rag.ts +6 -1
- package/packages/model-bank/src/aiModels/aihubmix.ts +74 -2
- package/packages/model-bank/src/aiModels/ollamacloud.ts +11 -0
- package/packages/model-bank/src/aiModels/openai.ts +74 -5
- package/packages/model-bank/src/aiModels/zenmux.ts +74 -0
- package/packages/model-runtime/src/const/models.ts +2 -0
- package/packages/model-runtime/src/core/contextBuilders/openai.ts +6 -2
- package/src/server/routers/lambda/image.ts +6 -6
- package/src/server/routers/lambda/user.ts +12 -4
package/CHANGELOG.md
CHANGED
|
@@ -2,6 +2,56 @@
|
|
|
2
2
|
|
|
3
3
|
# Changelog
|
|
4
4
|
|
|
5
|
+
## [Version 2.0.0-next.172](https://github.com/lobehub/lobe-chat/compare/v2.0.0-next.171...v2.0.0-next.172)
|
|
6
|
+
|
|
7
|
+
<sup>Released on **2025-12-15**</sup>
|
|
8
|
+
|
|
9
|
+
#### 💄 Styles
|
|
10
|
+
|
|
11
|
+
- **misc**: Update i18n.
|
|
12
|
+
|
|
13
|
+
<br/>
|
|
14
|
+
|
|
15
|
+
<details>
|
|
16
|
+
<summary><kbd>Improvements and Fixes</kbd></summary>
|
|
17
|
+
|
|
18
|
+
#### Styles
|
|
19
|
+
|
|
20
|
+
- **misc**: Update i18n, closes [#10759](https://github.com/lobehub/lobe-chat/issues/10759) ([24cae77](https://github.com/lobehub/lobe-chat/commit/24cae77))
|
|
21
|
+
|
|
22
|
+
</details>
|
|
23
|
+
|
|
24
|
+
<div align="right">
|
|
25
|
+
|
|
26
|
+
[](#readme-top)
|
|
27
|
+
|
|
28
|
+
</div>
|
|
29
|
+
|
|
30
|
+
## [Version 2.0.0-next.171](https://github.com/lobehub/lobe-chat/compare/v2.0.0-next.170...v2.0.0-next.171)
|
|
31
|
+
|
|
32
|
+
<sup>Released on **2025-12-14**</sup>
|
|
33
|
+
|
|
34
|
+
#### 💄 Styles
|
|
35
|
+
|
|
36
|
+
- **misc**: Update GPT-5.2 models.
|
|
37
|
+
|
|
38
|
+
<br/>
|
|
39
|
+
|
|
40
|
+
<details>
|
|
41
|
+
<summary><kbd>Improvements and Fixes</kbd></summary>
|
|
42
|
+
|
|
43
|
+
#### Styles
|
|
44
|
+
|
|
45
|
+
- **misc**: Update GPT-5.2 models, closes [#10749](https://github.com/lobehub/lobe-chat/issues/10749) ([0446127](https://github.com/lobehub/lobe-chat/commit/0446127))
|
|
46
|
+
|
|
47
|
+
</details>
|
|
48
|
+
|
|
49
|
+
<div align="right">
|
|
50
|
+
|
|
51
|
+
[](#readme-top)
|
|
52
|
+
|
|
53
|
+
</div>
|
|
54
|
+
|
|
5
55
|
## [Version 2.0.0-next.170](https://github.com/lobehub/lobe-chat/compare/v2.0.0-next.169...v2.0.0-next.170)
|
|
6
56
|
|
|
7
57
|
<sup>Released on **2025-12-12**</sup>
|
package/README.md
CHANGED
|
@@ -345,14 +345,14 @@ In addition, these plugins are not limited to news aggregation, but can also ext
|
|
|
345
345
|
|
|
346
346
|
<!-- PLUGIN LIST -->
|
|
347
347
|
|
|
348
|
-
| Recent Submits
|
|
349
|
-
|
|
|
350
|
-
| [
|
|
351
|
-
| [
|
|
352
|
-
| [
|
|
353
|
-
| [
|
|
354
|
-
|
|
355
|
-
> 📊 Total plugins: [<kbd>**
|
|
348
|
+
| Recent Submits | Description |
|
|
349
|
+
| ----------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------- |
|
|
350
|
+
| [Video Captions](https://lobechat.com/discover/plugin/VideoCaptions)<br/><sup>By **maila** on **2025-12-13**</sup> | Convert Youtube links into transcribed text, enable asking questions, create chapters, and summarize its content.<br/>`video-to-text` `youtube` |
|
|
351
|
+
| [WeatherGPT](https://lobechat.com/discover/plugin/WeatherGPT)<br/><sup>By **steven-tey** on **2025-12-13**</sup> | Get current weather information for a specific location.<br/>`weather` |
|
|
352
|
+
| [Git OSS Stats](https://lobechat.com/discover/plugin/gitUserRepoStats)<br/><sup>By **yunwei37** on **2025-12-13**</sup> | Dynamically generate and analyze stats and history for OSS repos and developers.<br/>`github` `oss` |
|
|
353
|
+
| [Questmate Forms](https://lobechat.com/discover/plugin/questmate)<br/><sup>By **questmate** on **2025-12-13**</sup> | Create forms, checklists and workflows (we call 'em Quests!) that you can assign, schedule or make public.<br/>`forms` `checklists` `productivity` |
|
|
354
|
+
|
|
355
|
+
> 📊 Total plugins: [<kbd>**41**</kbd>](https://lobechat.com/discover/plugins)
|
|
356
356
|
|
|
357
357
|
<!-- PLUGIN LIST -->
|
|
358
358
|
|
package/README.zh-CN.md
CHANGED
|
@@ -338,14 +338,14 @@ LobeChat 的插件生态系统是其核心功能的重要扩展,它极大地
|
|
|
338
338
|
|
|
339
339
|
<!-- PLUGIN LIST -->
|
|
340
340
|
|
|
341
|
-
| 最近新增
|
|
342
|
-
|
|
|
343
|
-
| [
|
|
344
|
-
| [
|
|
345
|
-
| [
|
|
346
|
-
| [
|
|
347
|
-
|
|
348
|
-
> 📊 Total plugins: [<kbd>**
|
|
341
|
+
| 最近新增 | 描述 |
|
|
342
|
+
| ----------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------- |
|
|
343
|
+
| [视频字幕](https://lobechat.com/discover/plugin/VideoCaptions)<br/><sup>By **maila** on **2025-12-13**</sup> | 将 Youtube 链接转换为转录文本,使其能够提问,创建章节,并总结其内容。<br/>`视频转文字` `you-tube` |
|
|
344
|
+
| [天气 GPT](https://lobechat.com/discover/plugin/WeatherGPT)<br/><sup>By **steven-tey** on **2025-12-13**</sup> | 获取特定位置的当前天气信息。<br/>`天气` |
|
|
345
|
+
| [Git OSS Stats](https://lobechat.com/discover/plugin/gitUserRepoStats)<br/><sup>By **yunwei37** on **2025-12-13**</sup> | 动态生成和分析开源软件仓库和开发者的统计数据和历史记录。<br/>`github` `oss` |
|
|
346
|
+
| [Questmate Forms](https://lobechat.com/discover/plugin/questmate)<br/><sup>By **questmate** on **2025-12-13**</sup> | 创建表单、清单和工作流程(我们称之为任务!),您可以分配、安排或公开。<br/>`表单` `清单` `生产力` |
|
|
347
|
+
|
|
348
|
+
> 📊 Total plugins: [<kbd>**41**</kbd>](https://lobechat.com/discover/plugins)
|
|
349
349
|
|
|
350
350
|
<!-- PLUGIN LIST -->
|
|
351
351
|
|
package/changelog/v1.json
CHANGED
|
@@ -1,4 +1,22 @@
|
|
|
1
1
|
[
|
|
2
|
+
{
|
|
3
|
+
"children": {
|
|
4
|
+
"improvements": [
|
|
5
|
+
"Update i18n."
|
|
6
|
+
]
|
|
7
|
+
},
|
|
8
|
+
"date": "2025-12-15",
|
|
9
|
+
"version": "2.0.0-next.172"
|
|
10
|
+
},
|
|
11
|
+
{
|
|
12
|
+
"children": {
|
|
13
|
+
"improvements": [
|
|
14
|
+
"Update GPT-5.2 models."
|
|
15
|
+
]
|
|
16
|
+
},
|
|
17
|
+
"date": "2025-12-14",
|
|
18
|
+
"version": "2.0.0-next.171"
|
|
19
|
+
},
|
|
2
20
|
{
|
|
3
21
|
"children": {},
|
|
4
22
|
"date": "2025-12-12",
|
|
@@ -125,6 +125,10 @@ table api_keys {
|
|
|
125
125
|
accessed_at "timestamp with time zone" [not null, default: `now()`]
|
|
126
126
|
created_at "timestamp with time zone" [not null, default: `now()`]
|
|
127
127
|
updated_at "timestamp with time zone" [not null, default: `now()`]
|
|
128
|
+
|
|
129
|
+
indexes {
|
|
130
|
+
user_id [name: 'api_keys_user_id_idx']
|
|
131
|
+
}
|
|
128
132
|
}
|
|
129
133
|
|
|
130
134
|
table async_tasks {
|
|
@@ -137,6 +141,10 @@ table async_tasks {
|
|
|
137
141
|
accessed_at "timestamp with time zone" [not null, default: `now()`]
|
|
138
142
|
created_at "timestamp with time zone" [not null, default: `now()`]
|
|
139
143
|
updated_at "timestamp with time zone" [not null, default: `now()`]
|
|
144
|
+
|
|
145
|
+
indexes {
|
|
146
|
+
user_id [name: 'async_tasks_user_id_idx']
|
|
147
|
+
}
|
|
140
148
|
}
|
|
141
149
|
|
|
142
150
|
table accounts {
|
|
@@ -215,6 +223,7 @@ table chat_groups {
|
|
|
215
223
|
|
|
216
224
|
indexes {
|
|
217
225
|
(client_id, user_id) [name: 'chat_groups_client_id_user_id_unique', unique]
|
|
226
|
+
group_id [name: 'chat_groups_group_id_idx']
|
|
218
227
|
}
|
|
219
228
|
}
|
|
220
229
|
|
|
@@ -303,6 +312,10 @@ table global_files {
|
|
|
303
312
|
creator text [not null]
|
|
304
313
|
created_at "timestamp with time zone" [not null, default: `now()`]
|
|
305
314
|
accessed_at "timestamp with time zone" [not null, default: `now()`]
|
|
315
|
+
|
|
316
|
+
indexes {
|
|
317
|
+
creator [name: 'global_files_creator_idx']
|
|
318
|
+
}
|
|
306
319
|
}
|
|
307
320
|
|
|
308
321
|
table knowledge_base_files {
|
|
@@ -313,6 +326,7 @@ table knowledge_base_files {
|
|
|
313
326
|
|
|
314
327
|
indexes {
|
|
315
328
|
(knowledge_base_id, file_id) [pk]
|
|
329
|
+
knowledge_base_id [name: 'knowledge_base_files_kb_id_idx']
|
|
316
330
|
}
|
|
317
331
|
}
|
|
318
332
|
|
|
@@ -332,6 +346,7 @@ table knowledge_bases {
|
|
|
332
346
|
|
|
333
347
|
indexes {
|
|
334
348
|
(client_id, user_id) [name: 'knowledge_bases_client_id_user_id_unique', unique]
|
|
349
|
+
user_id [name: 'knowledge_bases_user_id_idx']
|
|
335
350
|
}
|
|
336
351
|
}
|
|
337
352
|
|
|
@@ -349,6 +364,11 @@ table generation_batches {
|
|
|
349
364
|
accessed_at "timestamp with time zone" [not null, default: `now()`]
|
|
350
365
|
created_at "timestamp with time zone" [not null, default: `now()`]
|
|
351
366
|
updated_at "timestamp with time zone" [not null, default: `now()`]
|
|
367
|
+
|
|
368
|
+
indexes {
|
|
369
|
+
user_id [name: 'generation_batches_user_id_idx']
|
|
370
|
+
generation_topic_id [name: 'generation_batches_topic_id_idx']
|
|
371
|
+
}
|
|
352
372
|
}
|
|
353
373
|
|
|
354
374
|
table generation_topics {
|
|
@@ -359,6 +379,10 @@ table generation_topics {
|
|
|
359
379
|
accessed_at "timestamp with time zone" [not null, default: `now()`]
|
|
360
380
|
created_at "timestamp with time zone" [not null, default: `now()`]
|
|
361
381
|
updated_at "timestamp with time zone" [not null, default: `now()`]
|
|
382
|
+
|
|
383
|
+
indexes {
|
|
384
|
+
user_id [name: 'generation_topics_user_id_idx']
|
|
385
|
+
}
|
|
362
386
|
}
|
|
363
387
|
|
|
364
388
|
table generations {
|
|
@@ -372,6 +396,11 @@ table generations {
|
|
|
372
396
|
accessed_at "timestamp with time zone" [not null, default: `now()`]
|
|
373
397
|
created_at "timestamp with time zone" [not null, default: `now()`]
|
|
374
398
|
updated_at "timestamp with time zone" [not null, default: `now()`]
|
|
399
|
+
|
|
400
|
+
indexes {
|
|
401
|
+
user_id [name: 'generations_user_id_idx']
|
|
402
|
+
generation_batch_id [name: 'generations_batch_id_idx']
|
|
403
|
+
}
|
|
375
404
|
}
|
|
376
405
|
|
|
377
406
|
table message_chunks {
|
|
@@ -513,6 +542,7 @@ table messages {
|
|
|
513
542
|
session_id [name: 'messages_session_id_idx']
|
|
514
543
|
thread_id [name: 'messages_thread_id_idx']
|
|
515
544
|
agent_id [name: 'messages_agent_id_idx']
|
|
545
|
+
group_id [name: 'messages_group_id_idx']
|
|
516
546
|
}
|
|
517
547
|
}
|
|
518
548
|
|
|
@@ -731,6 +761,8 @@ table document_chunks {
|
|
|
731
761
|
|
|
732
762
|
indexes {
|
|
733
763
|
(document_id, chunk_id) [pk]
|
|
764
|
+
document_id [name: 'document_chunks_document_id_idx']
|
|
765
|
+
chunk_id [name: 'document_chunks_chunk_id_idx']
|
|
734
766
|
}
|
|
735
767
|
}
|
|
736
768
|
|
|
@@ -745,6 +777,7 @@ table embeddings {
|
|
|
745
777
|
indexes {
|
|
746
778
|
(client_id, user_id) [name: 'embeddings_client_id_user_id_unique', unique]
|
|
747
779
|
chunk_id [name: 'embeddings_chunk_id_idx']
|
|
780
|
+
user_id [name: 'embeddings_user_id_idx']
|
|
748
781
|
}
|
|
749
782
|
}
|
|
750
783
|
|
package/locales/ar/models.json
CHANGED
|
@@ -1229,6 +1229,9 @@
|
|
|
1229
1229
|
"deepseek_r1_distill_qwen_32b": {
|
|
1230
1230
|
"description": "DeepSeek-R1-Distill-Qwen-32B هو نموذج تم الحصول عليه من Qwen2.5-32B من خلال تقطير المعرفة. يستخدم هذا النموذج 800,000 عينة مختارة تم إنشاؤها بواسطة DeepSeek-R1 للتدريب، ويظهر أداءً ممتازًا في مجالات متعددة مثل الرياضيات، البرمجة، والاستدلال."
|
|
1231
1231
|
},
|
|
1232
|
+
"devstral-2:123b": {
|
|
1233
|
+
"description": "نموذج Devstral 2 123B، يتميز بالقدرة على استخدام الأدوات لاستكشاف قواعد الشيفرة، وتحرير ملفات متعددة، ودعم وكلاء هندسة البرمجيات."
|
|
1234
|
+
},
|
|
1232
1235
|
"doubao-1.5-lite-32k": {
|
|
1233
1236
|
"description": "دو باو 1.5 لايت هو نموذج الجيل الجديد الخفيف، مع سرعة استجابة قصوى، حيث يصل الأداء والوقت المستغرق إلى مستوى عالمي."
|
|
1234
1237
|
},
|
|
@@ -1937,6 +1940,15 @@
|
|
|
1937
1940
|
"gpt-5.1-codex-mini": {
|
|
1938
1941
|
"description": "GPT-5.1 Codex mini: إصدار مصغر ومنخفض التكلفة من Codex، مُحسَّن لمهام البرمجة القائمة على الوكلاء."
|
|
1939
1942
|
},
|
|
1943
|
+
"gpt-5.2": {
|
|
1944
|
+
"description": "GPT-5.2 — النموذج الرائد المخصص للبرمجة وتدفقات العمل الذكية، يتمتع بقدرات أقوى في الاستدلال وسياق طويل المدى."
|
|
1945
|
+
},
|
|
1946
|
+
"gpt-5.2-chat-latest": {
|
|
1947
|
+
"description": "GPT-5.2 Chat: إصدار GPT-5.2 المستخدم في ChatGPT (chat-latest)، لتجربة أحدث التحسينات في المحادثة."
|
|
1948
|
+
},
|
|
1949
|
+
"gpt-5.2-pro": {
|
|
1950
|
+
"description": "GPT-5.2 pro: إصدار أكثر ذكاءً ودقة من GPT-5.2 (لواجهة Responses API فقط)، مثالي للمسائل المعقدة والاستدلال متعدد الجولات الطويل."
|
|
1951
|
+
},
|
|
1940
1952
|
"gpt-audio": {
|
|
1941
1953
|
"description": "GPT Audio هو نموذج دردشة عام موجه لإدخال وإخراج الصوت، ويدعم استخدام الصوت في واجهة برمجة تطبيقات Chat Completions."
|
|
1942
1954
|
},
|
|
@@ -1229,6 +1229,9 @@
|
|
|
1229
1229
|
"deepseek_r1_distill_qwen_32b": {
|
|
1230
1230
|
"description": "DeepSeek-R1-Distill-Qwen-32B е модел, получен чрез знание на дестилация на Qwen2.5-32B. Този модел е финализиран с 800 000 избрани примера, генерирани от DeepSeek-R1, показвайки изключителна производителност в множество области, включително математика, програмиране и разсъждения."
|
|
1231
1231
|
},
|
|
1232
|
+
"devstral-2:123b": {
|
|
1233
|
+
"description": "Devstral 2 123B модел, специализиран в използването на инструменти за изследване на кодови бази, редактиране на множество файлове и подпомагане на агенти за софтуерно инженерство."
|
|
1234
|
+
},
|
|
1232
1235
|
"doubao-1.5-lite-32k": {
|
|
1233
1236
|
"description": "Doubao-1.5-lite е ново поколение лек модел, с изключителна скорост на отговор, който постига световно ниво както по отношение на ефективността, така и на времето за реакция."
|
|
1234
1237
|
},
|
|
@@ -1937,6 +1940,15 @@
|
|
|
1937
1940
|
"gpt-5.1-codex-mini": {
|
|
1938
1941
|
"description": "GPT-5.1 Codex mini: по-компактен и икономичен вариант на Codex, оптимизиран за агентски задачи по кодиране."
|
|
1939
1942
|
},
|
|
1943
|
+
"gpt-5.2": {
|
|
1944
|
+
"description": "GPT-5.2 — водещ модел за програмиране и агентски работни потоци, предлагащ по-силни способности за извеждане и работа с дълъг контекст."
|
|
1945
|
+
},
|
|
1946
|
+
"gpt-5.2-chat-latest": {
|
|
1947
|
+
"description": "GPT-5.2 Chat: Вариант на GPT-5.2, използван в ChatGPT (chat-latest), предназначен за най-новите подобрения в диалога."
|
|
1948
|
+
},
|
|
1949
|
+
"gpt-5.2-pro": {
|
|
1950
|
+
"description": "GPT-5.2 pro: По-умен и по-точен вариант на GPT-5.2 (само чрез Responses API), подходящ за сложни въпроси и по-дълги многоетапни разсъждения."
|
|
1951
|
+
},
|
|
1940
1952
|
"gpt-audio": {
|
|
1941
1953
|
"description": "GPT Audio е универсален чат модел, ориентиран към аудио вход и изход, поддържащ използване на аудио I/O в Chat Completions API."
|
|
1942
1954
|
},
|
|
@@ -1229,6 +1229,9 @@
|
|
|
1229
1229
|
"deepseek_r1_distill_qwen_32b": {
|
|
1230
1230
|
"description": "DeepSeek-R1-Distill-Qwen-32B ist ein Modell, das durch Wissensdistillation auf der Basis von Qwen2.5-32B entwickelt wurde. Dieses Modell wurde mit 800.000 ausgewählten Beispielen, die von DeepSeek-R1 generiert wurden, feinabgestimmt und zeigt in Mathematik, Programmierung und Schlussfolgerung in mehreren Bereichen herausragende Leistungen."
|
|
1231
1231
|
},
|
|
1232
|
+
"devstral-2:123b": {
|
|
1233
|
+
"description": "Devstral 2 123B Modell, spezialisiert auf den Einsatz von Tools zur Erkundung von Codebasen, Bearbeitung mehrerer Dateien und Unterstützung von Softwareentwicklungsagenten."
|
|
1234
|
+
},
|
|
1232
1235
|
"doubao-1.5-lite-32k": {
|
|
1233
1236
|
"description": "Doubao-1.5-lite ist das neueste leichte Modell der nächsten Generation, das eine extrem schnelle Reaktionszeit bietet und sowohl in der Leistung als auch in der Latenz weltweit erstklassig ist."
|
|
1234
1237
|
},
|
|
@@ -1937,6 +1940,15 @@
|
|
|
1937
1940
|
"gpt-5.1-codex-mini": {
|
|
1938
1941
|
"description": "GPT-5.1 Codex mini: Eine kompaktere und kostengünstigere Codex-Variante, optimiert für agentenbasierte Programmieraufgaben."
|
|
1939
1942
|
},
|
|
1943
|
+
"gpt-5.2": {
|
|
1944
|
+
"description": "GPT-5.2 — Das Flaggschiffmodell für Programmierung und agentenbasierte Workflows mit verbesserter Logik und erweitertem Kontextverständnis."
|
|
1945
|
+
},
|
|
1946
|
+
"gpt-5.2-chat-latest": {
|
|
1947
|
+
"description": "GPT-5.2 Chat: Die von ChatGPT verwendete GPT-5.2-Variante (chat-latest) für ein verbessertes Konversationserlebnis."
|
|
1948
|
+
},
|
|
1949
|
+
"gpt-5.2-pro": {
|
|
1950
|
+
"description": "GPT-5.2 pro: Eine intelligentere und präzisere Version von GPT-5.2 (nur über die Responses API verfügbar), ideal für komplexe Fragestellungen und längere, mehrstufige Schlussfolgerungen."
|
|
1951
|
+
},
|
|
1940
1952
|
"gpt-audio": {
|
|
1941
1953
|
"description": "GPT Audio ist ein universelles Chatmodell für Audioeingabe und -ausgabe, das Audio-I/O in der Chat Completions API unterstützt."
|
|
1942
1954
|
},
|
|
@@ -1229,6 +1229,9 @@
|
|
|
1229
1229
|
"deepseek_r1_distill_qwen_32b": {
|
|
1230
1230
|
"description": "DeepSeek-R1-Distill-Qwen-32B is a model derived from Qwen2.5-32B through knowledge distillation. This model is fine-tuned using 800,000 curated samples generated by DeepSeek-R1, demonstrating outstanding performance across multiple domains such as mathematics, programming, and reasoning."
|
|
1231
1231
|
},
|
|
1232
|
+
"devstral-2:123b": {
|
|
1233
|
+
"description": "Devstral 2 123B model, excels at using tools to explore codebases, edit multiple files, and support software engineering agents."
|
|
1234
|
+
},
|
|
1232
1235
|
"doubao-1.5-lite-32k": {
|
|
1233
1236
|
"description": "Doubao-1.5-lite is a new generation lightweight model, offering extreme response speed with performance and latency at a world-class level."
|
|
1234
1237
|
},
|
|
@@ -1937,6 +1940,15 @@
|
|
|
1937
1940
|
"gpt-5.1-codex-mini": {
|
|
1938
1941
|
"description": "GPT-5.1 Codex mini: A smaller, more cost-effective Codex variant optimized for agentic coding tasks."
|
|
1939
1942
|
},
|
|
1943
|
+
"gpt-5.2": {
|
|
1944
|
+
"description": "GPT-5.2 — A flagship model designed for coding and agentic workflows, offering enhanced reasoning and long-context capabilities."
|
|
1945
|
+
},
|
|
1946
|
+
"gpt-5.2-chat-latest": {
|
|
1947
|
+
"description": "GPT-5.2 Chat: The GPT-5.2 variant used by ChatGPT (chat-latest), featuring the latest improvements in conversational experience."
|
|
1948
|
+
},
|
|
1949
|
+
"gpt-5.2-pro": {
|
|
1950
|
+
"description": "GPT-5.2 Pro: A smarter, more precise version of GPT-5.2 (Responses API Only), ideal for complex problems and extended multi-turn reasoning."
|
|
1951
|
+
},
|
|
1940
1952
|
"gpt-audio": {
|
|
1941
1953
|
"description": "GPT Audio is a general-purpose chat model designed for audio input and output, supporting audio I/O in the Chat Completions API."
|
|
1942
1954
|
},
|
|
@@ -1229,6 +1229,9 @@
|
|
|
1229
1229
|
"deepseek_r1_distill_qwen_32b": {
|
|
1230
1230
|
"description": "DeepSeek-R1-Distill-Qwen-32B es un modelo obtenido a partir de Qwen2.5-32B mediante destilación de conocimiento. Este modelo se ajustó utilizando 800,000 muestras seleccionadas generadas por DeepSeek-R1, mostrando un rendimiento excepcional en múltiples campos como matemáticas, programación y razonamiento."
|
|
1231
1231
|
},
|
|
1232
|
+
"devstral-2:123b": {
|
|
1233
|
+
"description": "Modelo Devstral 2 123B, especializado en el uso de herramientas para explorar bases de código, editar múltiples archivos y asistir a agentes de ingeniería de software."
|
|
1234
|
+
},
|
|
1232
1235
|
"doubao-1.5-lite-32k": {
|
|
1233
1236
|
"description": "Doubao-1.5-lite es un modelo ligero de nueva generación, con una velocidad de respuesta extrema, alcanzando niveles de rendimiento y latencia de clase mundial."
|
|
1234
1237
|
},
|
|
@@ -1937,6 +1940,15 @@
|
|
|
1937
1940
|
"gpt-5.1-codex-mini": {
|
|
1938
1941
|
"description": "GPT-5.1 Codex mini: variante de Codex más compacta y económica, optimizada para tareas de codificación agentica."
|
|
1939
1942
|
},
|
|
1943
|
+
"gpt-5.2": {
|
|
1944
|
+
"description": "GPT-5.2 — Modelo insignia orientado a la codificación y flujos de trabajo agenticos, con capacidades mejoradas de razonamiento y contexto extenso."
|
|
1945
|
+
},
|
|
1946
|
+
"gpt-5.2-chat-latest": {
|
|
1947
|
+
"description": "GPT-5.2 Chat: Variante de GPT-5.2 utilizada por ChatGPT (chat-latest), diseñada para ofrecer las últimas mejoras en conversaciones."
|
|
1948
|
+
},
|
|
1949
|
+
"gpt-5.2-pro": {
|
|
1950
|
+
"description": "GPT-5.2 pro: Versión más inteligente y precisa de GPT-5.2 (solo API de respuestas), ideal para problemas complejos y razonamientos multietapa más largos."
|
|
1951
|
+
},
|
|
1940
1952
|
"gpt-audio": {
|
|
1941
1953
|
"description": "GPT Audio es un modelo de chat general para entrada y salida de audio, compatible con el uso de audio I/O en la API de Chat Completions."
|
|
1942
1954
|
},
|
|
@@ -1229,6 +1229,9 @@
|
|
|
1229
1229
|
"deepseek_r1_distill_qwen_32b": {
|
|
1230
1230
|
"description": "DeepSeek-R1-Distill-Qwen-32B مدلی است که بر اساس Qwen2.5-32B از طریق تقطیر دانش بهدست آمده است. این مدل از 800000 نمونه منتخب تولید شده توسط DeepSeek-R1 برای بهینهسازی استفاده میکند و در چندین حوزه از جمله ریاضی، برنامهنویسی و استدلال عملکرد فوقالعادهای دارد."
|
|
1231
1231
|
},
|
|
1232
|
+
"devstral-2:123b": {
|
|
1233
|
+
"description": "مدل Devstral 2 123B، متخصص در استفاده از ابزارها برای کاوش در مخازن کد، ویرایش چندین فایل و پشتیبانی از نمایندگان مهندسی نرمافزار."
|
|
1234
|
+
},
|
|
1232
1235
|
"doubao-1.5-lite-32k": {
|
|
1233
1236
|
"description": "مدل سبک نسل جدید Doubao-1.5-lite، با سرعت پاسخدهی فوقالعاده، عملکرد و تأخیر در سطح جهانی را ارائه میدهد."
|
|
1234
1237
|
},
|
|
@@ -1937,6 +1940,15 @@
|
|
|
1937
1940
|
"gpt-5.1-codex-mini": {
|
|
1938
1941
|
"description": "GPT-5.1 Codex mini: نسخهای کوچکتر و مقرونبهصرفهتر از Codex، بهینهسازیشده برای وظایف کدنویسی عاملمحور."
|
|
1939
1942
|
},
|
|
1943
|
+
"gpt-5.2": {
|
|
1944
|
+
"description": "GPT-5.2 — مدل پرچمدار برای برنامهنویسی و جریانهای کاری عاملمحور، با توانایی استدلال قویتر و پشتیبانی از زمینههای طولانیتر."
|
|
1945
|
+
},
|
|
1946
|
+
"gpt-5.2-chat-latest": {
|
|
1947
|
+
"description": "GPT-5.2 Chat: نسخهای از GPT-5.2 که در ChatGPT استفاده میشود (chat-latest)، برای تجربه بهبودهای جدید در مکالمه."
|
|
1948
|
+
},
|
|
1949
|
+
"gpt-5.2-pro": {
|
|
1950
|
+
"description": "GPT-5.2 pro: نسخهای هوشمندتر و دقیقتر از GPT-5.2 (فقط از طریق API پاسخها)، مناسب برای مسائل پیچیده و استدلالهای چندمرحلهای طولانی."
|
|
1951
|
+
},
|
|
1940
1952
|
"gpt-audio": {
|
|
1941
1953
|
"description": "GPT Audio مدلی عمومی برای چت با ورودی و خروجی صوتی است که از استفاده از ورودی/خروجی صوتی در API تکمیل چت پشتیبانی میکند."
|
|
1942
1954
|
},
|
|
@@ -1229,6 +1229,9 @@
|
|
|
1229
1229
|
"deepseek_r1_distill_qwen_32b": {
|
|
1230
1230
|
"description": "DeepSeek-R1-Distill-Qwen-32B est un modèle obtenu par distillation de connaissances basé sur Qwen2.5-32B. Ce modèle a été affiné à l'aide de 800 000 échantillons sélectionnés générés par DeepSeek-R1, montrant des performances exceptionnelles dans plusieurs domaines tels que les mathématiques, la programmation et le raisonnement."
|
|
1231
1231
|
},
|
|
1232
|
+
"devstral-2:123b": {
|
|
1233
|
+
"description": "Modèle Devstral 2 123B, spécialisé dans l'exploration de bases de code à l'aide d'outils, l'édition de fichiers multiples et le soutien aux agents en ingénierie logicielle."
|
|
1234
|
+
},
|
|
1232
1235
|
"doubao-1.5-lite-32k": {
|
|
1233
1236
|
"description": "Doubao-1.5-lite est un modèle léger de nouvelle génération, offrant une vitesse de réponse extrême, avec des performances et des délais atteignant des niveaux de classe mondiale."
|
|
1234
1237
|
},
|
|
@@ -1937,6 +1940,15 @@
|
|
|
1937
1940
|
"gpt-5.1-codex-mini": {
|
|
1938
1941
|
"description": "GPT-5.1 Codex mini : Variante plus légère et économique de Codex, optimisée pour les tâches de codage agentiques."
|
|
1939
1942
|
},
|
|
1943
|
+
"gpt-5.2": {
|
|
1944
|
+
"description": "GPT-5.2 — Modèle phare conçu pour le codage et les flux de travail agentiques, offrant des capacités de raisonnement avancées et une gestion étendue du contexte."
|
|
1945
|
+
},
|
|
1946
|
+
"gpt-5.2-chat-latest": {
|
|
1947
|
+
"description": "GPT-5.2 Chat : variante de GPT-5.2 utilisée par ChatGPT (chat-latest), permettant de découvrir les dernières améliorations en matière de conversation."
|
|
1948
|
+
},
|
|
1949
|
+
"gpt-5.2-pro": {
|
|
1950
|
+
"description": "GPT-5.2 pro : version plus intelligente et plus précise de GPT-5.2 (uniquement via l'API Responses), idéale pour les problèmes complexes et les raisonnements multi-tours prolongés."
|
|
1951
|
+
},
|
|
1940
1952
|
"gpt-audio": {
|
|
1941
1953
|
"description": "GPT Audio est un modèle de chat universel orienté vers l'entrée et la sortie audio, supportant l'utilisation d'entrées/sorties audio dans l'API Chat Completions."
|
|
1942
1954
|
},
|
|
@@ -1229,6 +1229,9 @@
|
|
|
1229
1229
|
"deepseek_r1_distill_qwen_32b": {
|
|
1230
1230
|
"description": "DeepSeek-R1-Distill-Qwen-32B è un modello ottenuto tramite distillazione della conoscenza basato su Qwen2.5-32B. Questo modello è stato affinato utilizzando 800.000 campioni selezionati generati da DeepSeek-R1, mostrando prestazioni eccezionali in vari campi, tra cui matematica, programmazione e ragionamento."
|
|
1231
1231
|
},
|
|
1232
|
+
"devstral-2:123b": {
|
|
1233
|
+
"description": "Devstral 2 123B, un modello specializzato nell'esplorazione di repository di codice tramite strumenti, nella modifica di più file e nel supporto agli agenti per l'ingegneria del software."
|
|
1234
|
+
},
|
|
1232
1235
|
"doubao-1.5-lite-32k": {
|
|
1233
1236
|
"description": "Doubao-1.5-lite è un modello leggero di nuova generazione, con una velocità di risposta eccezionale, raggiungendo standard di livello mondiale sia in termini di prestazioni che di latenza."
|
|
1234
1237
|
},
|
|
@@ -1937,6 +1940,15 @@
|
|
|
1937
1940
|
"gpt-5.1-codex-mini": {
|
|
1938
1941
|
"description": "GPT-5.1 Codex mini: Variante più compatta e conveniente di Codex, ottimizzata per compiti di codifica agentica."
|
|
1939
1942
|
},
|
|
1943
|
+
"gpt-5.2": {
|
|
1944
|
+
"description": "GPT-5.2 — modello di punta per la programmazione e i flussi di lavoro agentici, con capacità avanzate di ragionamento e gestione di contesti estesi."
|
|
1945
|
+
},
|
|
1946
|
+
"gpt-5.2-chat-latest": {
|
|
1947
|
+
"description": "GPT-5.2 Chat: variante di GPT-5.2 utilizzata da ChatGPT (chat-latest), progettata per offrire la più recente esperienza di conversazione migliorata."
|
|
1948
|
+
},
|
|
1949
|
+
"gpt-5.2-pro": {
|
|
1950
|
+
"description": "GPT-5.2 pro: una versione più intelligente e precisa di GPT-5.2 (solo tramite Responses API), ideale per problemi complessi e ragionamenti multi-turno più lunghi."
|
|
1951
|
+
},
|
|
1940
1952
|
"gpt-audio": {
|
|
1941
1953
|
"description": "GPT Audio è un modello di chat universale per input e output audio, supporta l'uso di I/O audio nell'API Chat Completions."
|
|
1942
1954
|
},
|
|
@@ -1229,6 +1229,9 @@
|
|
|
1229
1229
|
"deepseek_r1_distill_qwen_32b": {
|
|
1230
1230
|
"description": "DeepSeek-R1-Distill-Qwen-32BはQwen2.5-32Bに基づき、知識蒸留を通じて得られたモデルです。このモデルはDeepSeek-R1が生成した80万の選りすぐりのサンプルを使用して微調整され、数学、プログラミング、推論などの複数の分野で卓越した性能を示しています。"
|
|
1231
1231
|
},
|
|
1232
|
+
"devstral-2:123b": {
|
|
1233
|
+
"description": "Devstral 2 123B モデルは、ツールを活用してコードベースを探索し、複数のファイルを編集し、ソフトウェアエンジニアリングエージェントを支援することに優れています。"
|
|
1234
|
+
},
|
|
1232
1235
|
"doubao-1.5-lite-32k": {
|
|
1233
1236
|
"description": "Doubao-1.5-liteは全く新しい世代の軽量版モデルで、極限の応答速度を実現し、効果と遅延の両方で世界トップレベルに達しています。"
|
|
1234
1237
|
},
|
|
@@ -1937,6 +1940,15 @@
|
|
|
1937
1940
|
"gpt-5.1-codex-mini": {
|
|
1938
1941
|
"description": "GPT-5.1 Codex mini:より小型でコスト効率の高い Codex バリアントで、エージェント型のコーディングタスクに最適化。"
|
|
1939
1942
|
},
|
|
1943
|
+
"gpt-5.2": {
|
|
1944
|
+
"description": "GPT-5.2 — コーディングとエージェントワークフロー向けのフラッグシップモデルで、より強力な推論能力と長文コンテキスト処理に対応しています。"
|
|
1945
|
+
},
|
|
1946
|
+
"gpt-5.2-chat-latest": {
|
|
1947
|
+
"description": "GPT-5.2 Chat:ChatGPT が使用する GPT-5.2 のバリアント(chat-latest)で、最新の対話機能を体験できます。"
|
|
1948
|
+
},
|
|
1949
|
+
"gpt-5.2-pro": {
|
|
1950
|
+
"description": "GPT-5.2 pro:より賢く、より精密な GPT-5.2 のバージョン(Responses API のみ対応)で、難易度の高い問題や長いマルチターン推論に最適です。"
|
|
1951
|
+
},
|
|
1940
1952
|
"gpt-audio": {
|
|
1941
1953
|
"description": "GPT Audioは音声の入出力に対応した汎用チャットモデルで、Chat Completions APIでの音声I/O利用をサポートしています。"
|
|
1942
1954
|
},
|
|
@@ -1229,6 +1229,9 @@
|
|
|
1229
1229
|
"deepseek_r1_distill_qwen_32b": {
|
|
1230
1230
|
"description": "DeepSeek-R1-Distill-Qwen-32B는 Qwen2.5-32B를 기반으로 지식 증류를 통해 얻어진 모델입니다. 이 모델은 DeepSeek-R1이 생성한 80만 개의 선별된 샘플을 사용하여 미세 조정되어, 수학, 프로그래밍 및 추론 등 여러 분야에서 뛰어난 성능을 발휘합니다."
|
|
1231
1231
|
},
|
|
1232
|
+
"devstral-2:123b": {
|
|
1233
|
+
"description": "Devstral 2 123B 모델은 도구를 활용한 코드베이스 탐색, 다중 파일 편집 및 소프트웨어 엔지니어링 에이전트 지원에 능숙합니다."
|
|
1234
|
+
},
|
|
1232
1235
|
"doubao-1.5-lite-32k": {
|
|
1233
1236
|
"description": "Doubao-1.5-lite는 전혀 새로운 세대의 경량 모델로, 극한의 응답 속도를 자랑하며, 효과와 지연 모두 세계 최고 수준에 도달했습니다."
|
|
1234
1237
|
},
|
|
@@ -1937,6 +1940,15 @@
|
|
|
1937
1940
|
"gpt-5.1-codex-mini": {
|
|
1938
1941
|
"description": "GPT-5.1 Codex mini: 더 작고 비용 효율적인 Codex 변형으로, 에이전트 기반 코드 작업에 최적화되어 있습니다."
|
|
1939
1942
|
},
|
|
1943
|
+
"gpt-5.2": {
|
|
1944
|
+
"description": "GPT-5.2 — 코딩 및 에이전트 기반 워크플로우를 위한 플래그십 모델로, 향상된 추론 능력과 긴 문맥 처리 능력을 제공합니다."
|
|
1945
|
+
},
|
|
1946
|
+
"gpt-5.2-chat-latest": {
|
|
1947
|
+
"description": "GPT-5.2 Chat: ChatGPT에서 사용하는 GPT-5.2의 최신 대화형 변형(chat-latest)으로, 최신 대화 기능을 체험할 수 있습니다."
|
|
1948
|
+
},
|
|
1949
|
+
"gpt-5.2-pro": {
|
|
1950
|
+
"description": "GPT-5.2 pro: 더 똑똑하고 정밀한 GPT-5.2 버전(응답 API 전용)으로, 고난이도 문제 해결과 장기 다중 회차 추론에 적합합니다."
|
|
1951
|
+
},
|
|
1940
1952
|
"gpt-audio": {
|
|
1941
1953
|
"description": "GPT Audio는 오디오 입출력을 위한 범용 대화 모델로, Chat Completions API에서 오디오 I/O 사용을 지원합니다."
|
|
1942
1954
|
},
|
|
@@ -1229,6 +1229,9 @@
|
|
|
1229
1229
|
"deepseek_r1_distill_qwen_32b": {
|
|
1230
1230
|
"description": "DeepSeek-R1-Distill-Qwen-32B is een model dat is verkregen door kennisdistillatie van Qwen2.5-32B. Dit model is fijn afgesteld met 800.000 zorgvuldig geselecteerde voorbeelden gegenereerd door DeepSeek-R1, en toont uitstekende prestaties in verschillende domeinen zoals wiskunde, programmeren en redenering."
|
|
1231
1231
|
},
|
|
1232
|
+
"devstral-2:123b": {
|
|
1233
|
+
"description": "Devstral 2 123B-model, gespecialiseerd in het gebruiken van tools om codebases te verkennen, meerdere bestanden te bewerken en ondersteuning te bieden aan software-engineering agents."
|
|
1234
|
+
},
|
|
1232
1235
|
"doubao-1.5-lite-32k": {
|
|
1233
1236
|
"description": "Doubao-1.5-lite is de nieuwste generatie lichtgewicht model, met extreme responssnelheid en prestaties die wereldwijd tot de top behoren."
|
|
1234
1237
|
},
|
|
@@ -1937,6 +1940,15 @@
|
|
|
1937
1940
|
"gpt-5.1-codex-mini": {
|
|
1938
1941
|
"description": "GPT-5.1 Codex mini: Een compactere en kostenefficiëntere variant van Codex, geoptimaliseerd voor agentgerichte coderingstaken."
|
|
1939
1942
|
},
|
|
1943
|
+
"gpt-5.2": {
|
|
1944
|
+
"description": "GPT-5.2 — Het vlaggenschipmodel voor codering en agent-gebaseerde workflows, met verbeterde redeneercapaciteiten en ondersteuning voor lange contexten."
|
|
1945
|
+
},
|
|
1946
|
+
"gpt-5.2-chat-latest": {
|
|
1947
|
+
"description": "GPT-5.2 Chat: De GPT-5.2-variant die door ChatGPT wordt gebruikt (chat-latest), ontworpen om de nieuwste verbeteringen in conversaties te ervaren."
|
|
1948
|
+
},
|
|
1949
|
+
"gpt-5.2-pro": {
|
|
1950
|
+
"description": "GPT-5.2 pro: Een slimmer en nauwkeuriger versie van GPT-5.2 (alleen beschikbaar via de Responses API), geschikt voor complexe vraagstukken en langere, meerlagige redeneersessies."
|
|
1951
|
+
},
|
|
1940
1952
|
"gpt-audio": {
|
|
1941
1953
|
"description": "GPT Audio is een universeel chatmodel gericht op audio-invoer en -uitvoer, ondersteund in de Chat Completions API met audio I/O."
|
|
1942
1954
|
},
|
|
@@ -1229,6 +1229,9 @@
|
|
|
1229
1229
|
"deepseek_r1_distill_qwen_32b": {
|
|
1230
1230
|
"description": "DeepSeek-R1-Distill-Qwen-32B to model uzyskany poprzez destylację wiedzy z Qwen2.5-32B. Model ten został dostrojony przy użyciu 800 000 starannie wybranych próbek wygenerowanych przez DeepSeek-R1, wykazując doskonałe wyniki w wielu dziedzinach, w tym matematyce, programowaniu i wnioskowaniu."
|
|
1231
1231
|
},
|
|
1232
|
+
"devstral-2:123b": {
|
|
1233
|
+
"description": "Model Devstral 2 123B, specjalizuje się w używaniu narzędzi do eksploracji baz kodu, edytowaniu wielu plików oraz wspieraniu agentów inżynierii oprogramowania."
|
|
1234
|
+
},
|
|
1232
1235
|
"doubao-1.5-lite-32k": {
|
|
1233
1236
|
"description": "Doubao-1.5-lite to nowa generacja modelu o lekkiej konstrukcji, charakteryzująca się ekstremalną szybkością reakcji, osiągając światowy poziom zarówno w zakresie wydajności, jak i opóźnienia."
|
|
1234
1237
|
},
|
|
@@ -1937,6 +1940,15 @@
|
|
|
1937
1940
|
"gpt-5.1-codex-mini": {
|
|
1938
1941
|
"description": "GPT-5.1 Codex mini: mniejszy i tańszy wariant Codex, zoptymalizowany do zadań kodowania z udziałem agentów."
|
|
1939
1942
|
},
|
|
1943
|
+
"gpt-5.2": {
|
|
1944
|
+
"description": "GPT-5.2 — flagowy model zaprojektowany do kodowania i pracy agentowej, oferujący lepsze wnioskowanie i obsługę długiego kontekstu."
|
|
1945
|
+
},
|
|
1946
|
+
"gpt-5.2-chat-latest": {
|
|
1947
|
+
"description": "GPT-5.2 Chat: Wariant GPT-5.2 używany przez ChatGPT (chat-latest), zapewniający najnowsze ulepszenia w prowadzeniu rozmów."
|
|
1948
|
+
},
|
|
1949
|
+
"gpt-5.2-pro": {
|
|
1950
|
+
"description": "GPT-5.2 pro: inteligentniejsza i bardziej precyzyjna wersja GPT-5.2 (tylko API odpowiedzi), idealna do rozwiązywania trudnych problemów i dłuższych, wieloetapowych wnioskowań."
|
|
1951
|
+
},
|
|
1940
1952
|
"gpt-audio": {
|
|
1941
1953
|
"description": "GPT Audio to uniwersalny model konwersacyjny obsługujący wejście i wyjście audio, dostępny w API Chat Completions z obsługą audio I/O."
|
|
1942
1954
|
},
|
|
@@ -1229,6 +1229,9 @@
|
|
|
1229
1229
|
"deepseek_r1_distill_qwen_32b": {
|
|
1230
1230
|
"description": "DeepSeek-R1-Distill-Qwen-32B é um modelo obtido através da destilação de conhecimento do Qwen2.5-32B. Este modelo foi ajustado com 800 mil amostras selecionadas geradas pelo DeepSeek-R1, demonstrando desempenho excepcional em várias áreas, incluindo matemática, programação e raciocínio."
|
|
1231
1231
|
},
|
|
1232
|
+
"devstral-2:123b": {
|
|
1233
|
+
"description": "Modelo Devstral 2 123B, especializado no uso de ferramentas para explorar repositórios de código, editar múltiplos arquivos e oferecer suporte a agentes de engenharia de software."
|
|
1234
|
+
},
|
|
1232
1235
|
"doubao-1.5-lite-32k": {
|
|
1233
1236
|
"description": "Doubao-1.5-lite é a nova geração de modelo leve, com velocidade de resposta extrema, alcançando níveis de desempenho e latência de classe mundial."
|
|
1234
1237
|
},
|
|
@@ -1937,6 +1940,15 @@
|
|
|
1937
1940
|
"gpt-5.1-codex-mini": {
|
|
1938
1941
|
"description": "GPT-5.1 Codex mini: Variante do Codex com menor tamanho e custo reduzido, otimizada para tarefas de codificação com agentes."
|
|
1939
1942
|
},
|
|
1943
|
+
"gpt-5.2": {
|
|
1944
|
+
"description": "GPT-5.2 — modelo principal voltado para codificação e fluxos de trabalho com agentes, oferecendo raciocínio mais avançado e capacidade de contexto estendido."
|
|
1945
|
+
},
|
|
1946
|
+
"gpt-5.2-chat-latest": {
|
|
1947
|
+
"description": "GPT-5.2 Chat: variante do GPT-5.2 usada pelo ChatGPT (chat-latest), projetada para oferecer as melhorias mais recentes em conversação."
|
|
1948
|
+
},
|
|
1949
|
+
"gpt-5.2-pro": {
|
|
1950
|
+
"description": "GPT-5.2 pro: versão mais inteligente e precisa do GPT-5.2 (somente via API de Respostas), ideal para questões complexas e raciocínio prolongado em múltiplas etapas."
|
|
1951
|
+
},
|
|
1940
1952
|
"gpt-audio": {
|
|
1941
1953
|
"description": "GPT Audio é um modelo de chat universal voltado para entrada e saída de áudio, suportando uso de áudio I/O na API de Chat Completions."
|
|
1942
1954
|
},
|
|
@@ -1229,6 +1229,9 @@
|
|
|
1229
1229
|
"deepseek_r1_distill_qwen_32b": {
|
|
1230
1230
|
"description": "DeepSeek-R1-Distill-Qwen-32B — это модель, полученная на основе Qwen2.5-32B через дистилляцию знаний. Эта модель была дообучена на 800000 отобранных образцах, сгенерированных DeepSeek-R1, демонстрируя выдающиеся результаты в различных областях, включая математику, программирование и вывод."
|
|
1231
1231
|
},
|
|
1232
|
+
"devstral-2:123b": {
|
|
1233
|
+
"description": "Модель Devstral 2 123B, специализируется на использовании инструментов для изучения кодовой базы, редактировании нескольких файлов и поддержке программных агентов."
|
|
1234
|
+
},
|
|
1232
1235
|
"doubao-1.5-lite-32k": {
|
|
1233
1236
|
"description": "Doubao-1.5-lite - совершенно новое поколение легкой модели, с максимальной скоростью отклика, результаты и задержка достигают мирового уровня."
|
|
1234
1237
|
},
|
|
@@ -1937,6 +1940,15 @@
|
|
|
1937
1940
|
"gpt-5.1-codex-mini": {
|
|
1938
1941
|
"description": "GPT-5.1 Codex mini: более компактный и экономичный вариант Codex, оптимизированный для агентных задач программирования."
|
|
1939
1942
|
},
|
|
1943
|
+
"gpt-5.2": {
|
|
1944
|
+
"description": "GPT-5.2 — флагманская модель для программирования и агентных рабочих процессов, обладающая улучшенными возможностями рассуждения и работы с длинным контекстом."
|
|
1945
|
+
},
|
|
1946
|
+
"gpt-5.2-chat-latest": {
|
|
1947
|
+
"description": "GPT-5.2 Chat: вариант GPT-5.2, используемый в ChatGPT (chat-latest), предназначен для демонстрации последних улучшений в диалогах."
|
|
1948
|
+
},
|
|
1949
|
+
"gpt-5.2-pro": {
|
|
1950
|
+
"description": "GPT-5.2 pro: более умная и точная версия GPT-5.2 (только через Responses API), подходит для сложных задач и длительных многошаговых рассуждений."
|
|
1951
|
+
},
|
|
1940
1952
|
"gpt-audio": {
|
|
1941
1953
|
"description": "GPT Audio — универсальная чат-модель с поддержкой аудиовхода и аудиовыхода, доступная через API Chat Completions с аудио I/O."
|
|
1942
1954
|
},
|