27 commits

Author	SHA1	Message	Date
cotran	09a1bebff5	fix json formatting for tool call msg	2025-04-04 15:01:38 -07:00
Shuguang Chen	d8dabfeec0	Fix message formatting	2025-04-04 10:51:37 -07:00
Shuguang Chen	cbd181a092	Fix a bug in message formatting	2025-04-04 09:53:54 -07:00
co tran	0c3d52bfe4	format fix	2025-04-01 20:25:15 +00:00
co tran	1b39ee3dd8	fix bugs when no logprob for prefill and bug in function calling loop when it always no tool call	2025-04-01 20:18:56 +00:00
co tran	a3ab6be51c	modify hallucination threshold and temperature	2025-04-01 17:36:54 +00:00
Adil Hafeez	f2323f771c	update response from upstream llm to now include it in dict with "response"	2025-03-31 18:42:46 -07:00
co tran	5bd991e97b	fix precommit	2025-04-01 01:28:18 +00:00
co tran	b7916ce192	clean code + remove cmts	2025-04-01 01:19:29 +00:00
Shuguang Chen	6ec4c14407	Fix prompt prefilling	2025-03-31 15:08:38 -07:00
co tran	afe7cc9e9e	fix bug and test	2025-03-31 21:50:05 +00:00
co tran	cc0845bce4	fix hallucination loop	2025-03-31 21:08:47 +00:00
Shuguang Chen	f035d166c8	Fix hallucination check	2025-03-28 16:30:03 -07:00
Shuguang Chen	425f9b0dd5	Update model usage	2025-03-28 15:10:51 -07:00
Adil Hafeez	8290d1969f	use public endpoint for arch v1.1	2025-03-28 12:38:44 -07:00
CTran	a3f2b3cef9	add hallucination modification (#455) ... * add hallucination modification * disable test	2025-03-28 09:49:20 -07:00
Adil Hafeez	b31a7a569a	update rest and other parts of the code to work with arch fc 1.1	2025-03-28 03:04:21 -07:00
Shuguang Chen	8335f0c3de	minor update	2025-03-27 10:26:47 -07:00
Shuguang Chen	820c0443ee	disable hallucination check	2025-03-24 17:07:07 -07:00
Shuguang Chen	cf30e94415	Init update	2025-03-24 16:53:10 -07:00
Adil Hafeez	84cd1df7bf	add preliminary support for llm agents (#432)	2025-03-19 15:21:34 -07:00
Shuguang Chen	e77fc47225	Handle intent matching better in arch gateway (#391)	2025-03-04 12:49:13 -08:00
CTran	e7b370cd2f	fix error in function name + new thresholds (#406) ... * fix error in function name + new thresholds * fix * fix * remove example * remove example	2025-02-14 09:57:39 -08:00
Adil Hafeez	4ec03af16e	use archfc hosted on aws (#409)	2025-02-13 11:03:34 -08:00
Adil Hafeez	8de6eacfbd	spotify demo with optimized context window code change (#397)	2025-02-07 19:14:15 -08:00
Shuguang Chen	88a02dc478	Some fixes on model server (#362) ... * Some fixes on model server * Remove prompt_prefilling message * Fix logging * Fix poetry issues * Improve logging and update the support for text truncation * Fix tests * Fix tests * Fix tests * Fix modelserver tests * Update modelserver tests	2025-01-10 16:45:36 -08:00
Shuguang Chen	ba7279becb	Use intent model from archfc to pick prompt gateway (#328)	2024-12-20 13:25:01 -08:00