310 Commits

Author	SHA1	Message	Date
CaptainB	655ca53419	fix: refactor collation refresh and reindex logic to improve error handling and connection management	2025-08-19 15:49:22 +08:00
CaptainB	0e1ae663b1	fix: add migration to refresh collation and reindex database	2025-08-19 13:55:10 +08:00
shaohuzhang1	bc6a5a8869	fix: Non streaming sessions cannot count token consumption #3635 (#3709)	2025-07-22 18:43:10 +08:00
Mr.zhao	a9c46cd7e0	fix: Reorganize the model cache (#3315)	2025-06-19 18:01:45 +08:00
wxg0103	e12b1fe14e	refactor: gemini	2025-06-05 09:11:06 +08:00
wxg0103	81a3af2c8b	refactor: openai token	2025-05-22 09:11:12 +08:00
wxg0103	2ec0d22b14	refactor: openai token	2025-05-21 18:50:49 +08:00
wxg0103	27a77dc657	refactor: openai token	2025-05-21 18:33:56 +08:00
wxg0103	76ec8ad6f6	refactor: regolo icon	2025-05-07 10:51:43 +08:00
Daniele Scasciafratte	0cf05a76a0	feat(provider): added regolo.ai (#3041)	2025-05-07 10:12:47 +08:00
shaohuzhang1	79b2de8893	feat: Ali Bailian supplier model list adds qwen3 model (#3026)	2025-04-29 19:08:17 +08:00
shaohuzhang1	a6533c0db7	feat: Support Alibaba Bailian supplier qwen3 model (#3020)	2025-04-29 14:57:41 +08:00
shaohuzhang1	33b1cd65b0	fix: Ollama maximum output token field (#2973)	2025-04-24 16:16:36 +08:00
shaohuzhang1	b1f6092620	fix: The tag based thinking process does not output (#2968)	2025-04-24 10:57:10 +08:00
wxg0103	e58f95832b	fix: aws computer tokens error ... --bug=1055094 --user=王孝刚 【应用】aws模型在应用编排中使用时，回答内容会带一串报错信息，并且消耗tokens显示为0 https://www.tapd.cn/57709429/s/1689448	2025-04-23 10:41:18 +08:00
shaohuzhang1	d6f1d25b59	fix: Change the max_token of the Qianfan large model to max_output_token (#2955)	2025-04-22 18:10:49 +08:00
shaohuzhang1	d2637c3de2	fix: Model parameters are not effective (#2937)	2025-04-21 18:06:09 +08:00
wxg0103	a4f27249ed	fix: openai error	2025-04-10 14:41:05 +08:00
wxg0103	add9d1bab8	fix: In the permission settings, the knowledge base list does not distinguish between knowledge base types, and all are marked with the icon of the general knowledge base. ... --bug=1054142 --user=王孝刚 【团队成员】权限设置中，知识库列表没有区分知识库类型，都用通用知识库的图标标识 https://www.tapd.cn/57709429/s/1681104	2025-04-07 13:53:39 +08:00
shaohuzhang1	867c53984b	fix: The local rearrangement model cannot be loaded (#2805)	2025-04-07 11:22:01 +08:00
shaohuzhang1	9d6451b95b	fix: The interface for obtaining model metadata cannot access public models (#2787)	2025-04-02 20:10:20 +08:00
shaohuzhang1	add1cba8cb	feat: Swagger document response for adding OpenAI interface (#2786)	2025-04-02 19:59:57 +08:00
shaohuzhang1	ac5a9d01a8	fix: When other users download the public model, they cannot obtain the download data (#2785)	2025-04-02 18:29:25 +08:00
CaptainB	7b213f547d	fix: update default voice type and options in VolcanicEngineTTSModelGeneralParams	2025-04-02 18:25:06 +08:00
shaohuzhang1	15feca802a	fix: OpenAI Vector Model Using Openai Supplier (#2781)	2025-04-02 17:45:18 +08:00
wxg0103	2686e76c8a	fix: xinference rerank error ... --bug=1054256 --user=王孝刚 【模型】添加硅基流动的重排序模型失败 https://www.tapd.cn/57709429/s/1679612	2025-04-02 16:54:58 +08:00
wxg0103	44c1d35b1f	fix: siliconCloud rerank error ... --bug=1054256 --user=王孝刚 【模型】添加硅基流动的重排序模型失败 https://www.tapd.cn/57709429/s/1679612	2025-04-02 15:17:39 +08:00
wxg0103	6d7b5eb219	fix: Increase log output	2025-04-02 14:17:48 +08:00
wxg0103	6fde8ec80f	fix: openai need streaming	2025-04-01 16:58:08 +08:00
wxg0103	a2b6620b10	refactor: azure llm params	2025-03-31 15:31:52 +08:00
wxg0103	2fe4248785	refactor: qwq model	2025-03-28 14:05:33 +08:00
wxg0103	c66f79ad5a	refactor: log menu	2025-03-27 16:14:24 +08:00
wxg0103	f82ba3c4b8	fix: qwq-plus only supported stream	2025-03-27 10:18:58 +08:00
CaptainB	1e8e3a90aa	fix: handle audio generation failure by raising an exception	2025-03-26 17:15:23 +08:00
CaptainB	d27e5a4c01	chore: fix warning	2025-03-26 13:35:29 +08:00
shaohuzhang1	f9d536f5a2	feat: Audit log add operation object (#2681)	2025-03-25 19:07:02 +08:00
wxg0103	2faabbe392	refactor: bailian	2025-03-21 11:02:43 +08:00
CaptainB	72398408c5	ci: update langchain version	2025-03-20 18:55:43 +08:00
shaohuzhang1	470105f895	feat: Support application、setting、user log recorder (#2617)	2025-03-19 15:45:10 +08:00
shaohuzhang1	be31989ab9	fix: Log annotation acquisition details file type data cannot be serialized (#2579)	2025-03-17 16:45:06 +08:00
shaohuzhang1	d7ef3ba67b	feat: Audit Log	2025-03-17 15:34:41 +08:00
shaohuzhang1	ea9c5e0ee8	fix: Model addition, error log printing (#2499)	2025-03-05 14:45:50 +08:00
CaptainB	801911d765	refactor: update QwenChatModel to use BaseChatOpenAI and remove unused methods ... --bug=1052269 --user=刘瑞斌 【模型】对接千问模型，设置联网参数，maxkb对答不生效。通过python代码调用是可以的。 https://www.tapd.cn/57709429/s/1662132	2025-03-04 17:52:37 +08:00
CaptainB	7e3f631dfc	refactor: add model_name parameter to AzureChatModel initialization ... --bug=1052553 --user=刘瑞斌 【模型】添加 Azure OpenAI 添加DeepSeek -R1 模型报错 https://www.tapd.cn/57709429/s/1661953	2025-03-04 16:08:20 +08:00
CaptainB	13ce64e51a	fix: Compatible with dashscope tts ... --bug=1052708 --user=刘瑞斌 【应用】-使用阿里云语音合成模型sambert-zhichu-v1，对话报错 https://www.tapd.cn/57709429/s/1661551	2025-03-04 10:49:36 +08:00
itaa	ed8f8f8a3e	fix: ollama model provider can not set 'num_ctx' etc. parameter #2442 (#2444) ... langchain-openai is not compatible with parameter Settings in ollama, such as num_ctx. Therefore, you need to create model instances using langchain-ollama (cherry picked from commit 42ae7b443d)	2025-02-28 17:45:48 +08:00
shaohuzhang1	eb6f4e8cb8	fix: VLLM Supplier Dialogue Verification (#2412)	2025-02-26 12:04:25 +08:00
shaohuzhang1	e1f0f39987	fix: When adding members in bulk, existing members will be automatically passed in, do not affect the joining of other members #2351 (#2397)	2025-02-25 16:54:29 +08:00
wxg0103	a01f21e3ac	refactor: ollama reranker	2025-02-25 11:07:49 +08:00
wxg0103	9185515660	refactor: ollama support rerank	2025-02-24 18:35:37 +08:00