vominh1919
5401a0080d
fix: recalculate token budgets on model switch in ContextCompressor
...
update_model() recalculated threshold_tokens but left tail_token_budget
and max_summary_tokens at their __init__ values. When switching from a
200K model to 32K, the tail budget stayed at ~20K tokens (62% of 32K)
instead of the intended ~10%.
Adds budget recalculation in update_model() and 2 regression tests.
2026-04-25 15:07:56 +05:30
..
2026-04-23 16:11:57 -07:00
2026-02-21 22:31:43 -08:00
2026-04-21 01:56:35 -07:00
2026-04-24 07:26:07 -07:00
2026-04-24 07:26:07 -07:00
2026-04-24 07:26:07 -07:00
2026-04-24 14:39:59 -07:00
2026-04-25 15:07:56 +05:30
2026-04-24 02:55:43 -07:00
2026-04-20 01:56:41 -07:00
2026-04-24 05:09:08 -07:00
2026-04-24 05:20:05 -07:00
2026-04-21 01:52:49 -07:00
2026-04-19 22:45:47 -07:00
2026-04-24 05:35:43 -07:00
2026-04-21 01:31:58 -07:00
2026-04-21 00:50:58 -07:00
2026-04-24 05:35:17 -07:00
2026-04-24 03:40:00 -07:00
2026-04-17 15:34:12 -07:00
2026-04-16 16:49:00 -07:00
2026-04-21 21:30:10 -07:00
2026-04-21 21:30:10 -07:00
2026-04-20 05:25:49 -07:00
2026-04-10 21:16:53 -07:00
2026-04-24 14:37:55 -07:00
2026-04-24 14:37:55 -07:00
2026-04-24 07:26:07 -07:00
2026-04-23 01:59:20 -07:00
2026-04-23 16:11:57 -07:00
2026-04-15 16:31:48 -07:00
2026-04-23 12:50:22 +05:30
2026-03-21 16:54:43 -07:00
2026-04-13 16:32:04 -07:00
2026-04-20 11:49:54 -07:00
2026-04-08 00:41:36 -07:00
2026-04-20 20:53:51 -07:00
2026-04-24 15:15:07 -07:00
2026-04-24 15:15:07 -07:00
2026-04-22 17:43:30 -07:00
2026-04-09 03:10:30 -07:00
2026-04-22 18:44:07 -07:00
2026-02-21 22:31:43 -08:00
2026-04-22 17:40:49 -07:00