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
..
2026-09-16 14:59:23 -07:00
2026-09-24 20:06:57 -07:00
2026-08-28 11:18:41 -07:00
2026-09-16 07:46:23 -07:00
2026-08-28 02:20:06 -07:00
2026-09-16 07:46:23 -07:00
2026-09-26 19:45:28 -07:00
2026-09-21 19:55:00 -07:00
2026-08-13 20:19:41 -07:00
2026-09-21 12:46:02 -07:00
2026-09-26 19:25:46 -07:00
2026-09-26 19:45:28 -07:00
2026-09-16 07:43:33 -07:00
2026-09-16 07:33:38 -07:00
2026-08-13 20:19:41 -07:00
2026-10-01 06:55:48 -07:00
2026-10-01 13:05:16 -07:00
2026-10-01 06:55:48 -07:00
2026-09-04 12:54:28 -07:00
2026-08-22 15:49:44 -07:00
2026-09-16 19:28:34 -07:00
2026-09-21 14:12:38 -07:00
2026-09-21 13:20:15 -07:00
2026-09-26 19:45:28 -07:00
2026-08-25 20:13:33 -07:00
2026-09-21 15:02:14 -07:00
2026-08-22 15:49:44 -07:00
2026-09-03 16:15:43 -07:00
2026-09-12 10:09:12 -07:00
2026-08-25 15:38:57 -07:00
2026-09-26 19:45:28 -07:00
2026-09-24 20:39:28 -07:00
2026-08-26 20:08:30 -07:00
2026-08-13 20:19:41 -07:00
2026-09-16 06:59:44 -07:00
2026-09-26 23:25:55 -07:00
2026-08-19 00:00:42 -07:00
2026-09-01 14:40:10 -07:00