Frank
148a5f2575
frank-primary: wire --decode-context-parallel-size (opt-in)
...
DCP is passed only when FRANK_PRIMARY_DCP is set, so the deployed shape is
unchanged. Verified in this fork: DCP reuses the TP ranks when PCP is
disabled (tp % dcp == 0), so TP=2 + DCP=2 fits the two R9700s with no extra
ranks; the flash_attn backend carries DCP support, the 36 linear-attention
layers of this hybrid do not — which is what the A/B window tests.
2026-10-01 13:05:16 -07:00
..
2026-09-20 18:15:27 -07:00
2026-09-30 20:13:23 -07:00
2026-08-28 02:20:06 -07:00
2026-09-16 21:43:09 -07:00
2026-09-20 17:51:27 -07:00
2026-09-03 16:15:43 -07:00
2026-09-16 06:59:44 -07:00
2026-09-03 16:15:43 -07:00
2026-09-03 16:15:43 -07:00
2026-09-16 15:38:20 -07:00
2026-09-03 16:15:43 -07:00
2026-09-28 22:08:29 -07:00
2026-09-21 18:56:35 -07:00
2026-09-20 15:25:09 -07:00
2026-09-16 07:43:33 -07:00
2026-09-20 15:25:09 -07:00
2026-09-28 23:27:50 -07:00
2026-08-25 20:13:33 -07:00
2026-08-25 20:13:33 -07:00
2026-09-26 19:45:28 -07:00
2026-10-01 06:55:48 -07:00
2026-08-25 20:13:33 -07:00
2026-09-26 19:25:46 -07:00
2026-09-19 01:42:37 -07:00
2026-09-28 22:48:53 -07:00
2026-09-21 13:58:54 -07:00
2026-09-21 21:50:36 -07:00
2026-09-30 20:13:23 -07:00
2026-10-01 06:55:48 -07:00
2026-09-26 19:51:57 -07:00
2026-09-28 23:27:50 -07:00
2026-09-28 22:08:29 -07:00
2026-09-26 21:58:03 -07:00
2026-10-01 04:14:25 -07:00
2026-10-01 04:14:25 -07:00
2026-10-01 06:55:48 -07:00
2026-10-01 07:06:15 -07:00
2026-08-28 02:20:06 -07:00
2026-09-26 15:23:24 -07:00
2026-09-26 22:50:17 -07:00
2026-09-21 13:20:15 -07:00
2026-08-23 12:59:56 -07:00
2026-08-23 12:59:56 -07:00
2026-08-25 15:38:57 -07:00
2026-09-26 19:25:46 -07:00
2026-09-26 23:25:55 -07:00
2026-09-28 23:27:50 -07:00
2026-10-01 13:05:16 -07:00
2026-10-01 06:55:48 -07:00
2026-09-17 18:38:54 -07:00
2026-09-17 13:09:59 -07:00
2026-09-26 19:45:28 -07:00
2026-08-25 17:04:15 -07:00