exo/lm-lib
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
57 Commits 1 Branch 0 Tags
70f7a06da8b9069dbf0db1c970f5b32e4db35e95
Commit Graph

42 Commits

Author SHA1 Message Date
extio-de
70f7a06da8 refactor: Remove unused import for Value annotation in AzureAiClientAutoConfiguration 2025-07-13 10:34:11 +02:00
extio-de
d0b256502b Use netty client for Azure, remove proxy configuration 2025-07-13 10:33:17 +02:00
extio-de
d9fae7eebd chore: Update version to 1.6.6 and refactor agentType method for context handling 2025-07-10 00:20:16 +02:00
extio-de
df3034e6ca Moved TextUtils to lm-lib 2025-07-05 20:04:56 +02:00
extio-de
7cf067400b feat: Allow AgentContext update consumer in BaseAgent and add beforeStream to TextAgentResponseHandler 
This commit introduces the following changes:

- **BaseAgent:** Added a mechanism to allow consumers to update the AgentContext after post-processing a split. This provides more flexibility for downstream components to react to the processed data.
- **TextAgentResponseHandler:** Introduced a `beforeStream` consumer that allows updating the AgentContext before the stream begins.  This enables pre-processing or initialization steps within the response handler.  The constructor has been updated to accept this new consumer.  Also fixed a bug where the chunkKey and updateKey were not being removed after processing.

These changes enhance the extensibility and configurability of the agent processing pipeline.
 2025-07-02 21:45:44 +02:00
extio-de
c5e2ace604 Fix TextAgentResponseHandler, Added afterChunkUpdate() 2025-06-29 17:29:37 +02:00
extio-de
9ca590a8d4 Implemented agent streaming 2025-06-28 16:55:26 +02:00
extio-de
365043b5a0 Streaming Client 2025-06-28 14:36:26 +02:00
extio-de
2350740136 Introduced OpenAI chat completion client; 2025-06-27 23:39:17 +02:00
extio-de
eee37a5648 Rename oai client package 2025-06-27 21:49:30 +02:00
extio-de
a555d00f74 Get token usage from OpenAI Api if available 2025-06-27 21:19:26 +02:00
extio-de
47ef9ba436 Refactored llm clients, removed maxContinuations feature, exposed finishReason instead; 2025-06-27 20:43:04 +02:00
extio-de
bef3ef1468 Deprecated Client#completion and removed redundant question parameter 2025-06-27 17:35:48 +02:00
extio-de
5b5f56fa15 Combined unnecessary dual prompt template parameters instruction and text to user 2025-06-27 17:17:27 +02:00
extio-de
4e52c1c4ce Code cleanup 2025-06-27 16:59:52 +02:00
extio-de
60bdc54f96 Moved oai client dtos to domain packages 2025-06-27 16:49:37 +02:00
extio-de
11a4ab1271 Moved oai client packages 2025-06-27 16:46:56 +02:00
extio-de
6595075846 Fix AzureAi Auto Config - Port default value 2025-06-27 16:46:29 +02:00
extio-de
94894aee85 Merged with tp lmlib 2025-06-27 16:36:46 +02:00
extio-de
dcd6eeee41 Refactor BaseAgent 2025-06-27 10:25:54 +02:00
extio-de
38138b177b Agent: Skip branching in case of empty template inputs; 2025-06-26 00:19:35 +02:00
extio-de
c3fd40f4a7 AgentNext.END constant and bump version to 1.5.6 2025-06-25 20:20:50 +02:00
extio-de
d76e2e8f4f Added Grader to library 2025-06-23 20:16:23 +02:00
extio-de
cf6b868d3c Update version to 1.5.4 and enhance BaseAgent's chooseNext method 2025-06-23 13:31:49 +02:00
extio-de
52963e4eed AgentContext: Convenience value getter and setter; 2025-06-23 12:48:59 +02:00
extio-de
1ab2340714 Agent walkGraph convenience overload; 2025-06-23 12:42:40 +02:00
extio-de
8aee0acb15 AgentContext new ctor; 2025-06-23 12:37:16 +02:00
extio-de
2378e8c758 Moved Agent implementation to BaseAgent interface 2025-06-23 09:09:24 +02:00
extio-de
dc455165c8 Agent: skipNextCompletion; Better cache keys; XL category; Fix ModelProfileService concurrency; 2025-06-22 13:28:22 +02:00
(no author)
04cde2558a Merged lm-lib 
git-svn-id: svn://home0/lm-lib@3737 71dcc7fd-b1f2-4e5d-bb8e-274146e81173
 2025-04-15 17:51:39 +00:00
(no author)
776ab01e85 lm-lib: Merged with tp fork; 
git-svn-id: svn://home0/lm-lib@3725 71dcc7fd-b1f2-4e5d-bb8e-274146e81173
 2025-02-23 20:22:10 +00:00
(no author)
3c412077bb lm-lib: Merged with tp fork; 
git-svn-id: svn://home0/lm-lib@3723 71dcc7fd-b1f2-4e5d-bb8e-274146e81173
 2025-02-23 19:08:49 +00:00
(no author)
c5ada1049d Phi4 prompt template; 
git-svn-id: svn://home0/lm-lib@3685 71dcc7fd-b1f2-4e5d-bb8e-274146e81173
 2025-01-11 22:21:35 +00:00
(no author)
fbdba7d76d Agents: {{{xxx}}} placeholders to insert multi values in json format; AgentContext: Convenience methods; Agents: Better error handling; Updated prompt strategies; More response handlers; 
git-svn-id: svn://home0/lm-lib@3653 71dcc7fd-b1f2-4e5d-bb8e-274146e81173
 2024-12-22 22:51:01 +00:00
(no author)
cb2c1c02da Calculate effective duration and tps; 
git-svn-id: svn://home0/lm-lib@3652 71dcc7fd-b1f2-4e5d-bb8e-274146e81173
 2024-12-22 12:34:07 +00:00
(no author)
63087a1cd6 More resilient json parsing; Removed responseAsUserPrompt feature; Better error handling; chooseNextAgent() is called after merger(); apply templates: multiple [[]] branched placeholders are now possible; New AgentType.CONVERSATION_WITH_SYSTEM_PROMPT; 
git-svn-id: svn://home0/lm-lib@3644 71dcc7fd-b1f2-4e5d-bb8e-274146e81173
 2024-12-21 14:45:53 +00:00
(no author)
1976daf697 Fix Alpaca prompt template; 
git-svn-id: svn://home0/lm-lib@3639 71dcc7fd-b1f2-4e5d-bb8e-274146e81173
 2024-12-19 23:12:37 +00:00
(no author)
170b7074e1 lm-lib ModelProfileService only query distinct urls; 
git-svn-id: svn://home0/lm-lib@3613 71dcc7fd-b1f2-4e5d-bb8e-274146e81173
 2024-12-18 19:18:47 +00:00
(no author)
f01309526a lm-lib refactoring; 
git-svn-id: svn://home0/lm-lib@3610 71dcc7fd-b1f2-4e5d-bb8e-274146e81173
 2024-12-18 18:58:27 +00:00
(no author)
35b59043e1 lm-lib fixes; 
git-svn-id: svn://home0/lm-lib@3609 71dcc7fd-b1f2-4e5d-bb8e-274146e81173
 2024-12-18 18:39:04 +00:00
(no author)
d3797e4d6d Fixes; 
git-svn-id: svn://home0/lm-lib@3605 71dcc7fd-b1f2-4e5d-bb8e-274146e81173
 2024-12-16 23:30:01 +00:00
(no author)
c1a283305d lm-lib first commit; 
git-svn-id: svn://home0/lm-lib@3603 71dcc7fd-b1f2-4e5d-bb8e-274146e81173
 2024-12-16 20:58:04 +00:00