Frank
c97e359bc2
docs: the LLM index and the cooperation page catch up with layer split
...
The LLM index still opened with vllm-radiance as frank-medium's runtime and
drew the replica pair in its topology diagram; the cooperation page's thesis
(collectives are impossible, so the answer is hardware) now has to name the
path the fleet actually took: llama.cpp's layer split, which launches no RCCL
kernel on either card. That reframes the M.2 riser as a means to TENSOR
parallelism rather than a means to using both cards at all.
2026-09-16 15:39:47 -07:00
..
2026-09-03 16:17:14 -07:00
2026-09-03 16:17:14 -07:00
2026-09-03 16:17:14 -07:00
2026-09-03 16:17:14 -07:00
2026-09-03 16:17:14 -07:00
2026-09-11 14:48:28 -07:00
2026-09-15 19:27:04 -07:00
2026-09-03 16:17:14 -07:00
2026-09-03 16:17:14 -07:00
2026-09-16 14:52:51 -07:00
2026-09-03 16:17:14 -07:00
2026-09-03 16:17:14 -07:00
2026-09-16 15:39:47 -07:00
2026-09-03 16:17:14 -07:00
2026-09-03 16:17:14 -07:00
2026-09-03 16:17:14 -07:00
2026-09-16 14:52:51 -07:00
2026-09-16 15:38:20 -07:00
2026-09-16 15:15:02 -07:00
2026-09-16 15:15:02 -07:00
2026-09-11 14:48:28 -07:00
2026-09-03 16:17:14 -07:00
2026-09-03 16:17:14 -07:00
2026-09-16 14:52:51 -07:00
2026-09-03 16:17:14 -07:00
2026-09-15 19:27:04 -07:00
2026-09-16 14:52:51 -07:00
2026-08-28 02:20:06 -07:00
2026-09-03 16:17:14 -07:00
2026-09-16 15:39:47 -07:00
2026-09-03 16:17:14 -07:00
2026-09-15 19:27:04 -07:00
2026-09-03 16:17:14 -07:00
2026-09-11 14:48:28 -07:00
2026-09-03 16:17:14 -07:00
2026-09-03 16:17:14 -07:00
2026-08-28 02:20:06 -07:00
2026-09-03 16:17:14 -07:00
2026-09-03 16:17:14 -07:00
2026-08-28 02:20:06 -07:00
2026-09-03 16:17:14 -07:00
2026-09-03 16:17:14 -07:00
2026-08-28 02:20:06 -07:00
2026-09-03 16:17:14 -07:00
2026-08-28 02:20:06 -07:00
2026-08-28 02:20:06 -07:00
2026-08-28 02:20:06 -07:00