分支列表

默认分支

e6be57ff4b · fix(comm): dispatch the one-shot all-reduce on the call's width (#1488) · 最后更新于 2026-09-11 02:55:57 +00:00

分支列表

fc75f40bfb · Keep the L3 PR focused by removing unrelated CI changes · 最后更新于 2026-09-11 03:57:17 +00:00    frostbyte_neo

0
79

6fc9da5c5e · ci: retry unsuccessful Slurm cases at the original commit · 最后更新于 2026-09-11 03:17:08 +00:00    frostbyte_neo

2
1

da42598442 · feat: support DeepSeek V4.1 text inference · 最后更新于 2026-09-11 03:04:15 +00:00    frostbyte_neo

20
1

c225c9842e · Merge main into Qwen4-Exp backend composition · 最后更新于 2026-09-11 00:53:24 +00:00    frostbyte_neo

2
27

9451891ad0 · Add AMD kernel benchmark regression CI · 最后更新于 2026-09-10 23:07:30 +00:00    frostbyte_neo

2
2

843fac0ef6 · fmt · 最后更新于 2026-09-10 20:19:50 +00:00    frostbyte_neo

13
15

0eab9e0366 · perf(amd): enable GLM-5.3 FP8 dense prefill · 最后更新于 2026-09-10 17:45:56 +00:00    frostbyte_neo

13
1

3d9dbd825e · perf(comm): tune TP4 Iris decode launches · 最后更新于 2026-09-10 17:41:22 +00:00    frostbyte_neo

13
6

eb0ecd7799 · perf(mhc): tune GLM-5.3-Flash reduction grid · 最后更新于 2026-09-10 16:59:16 +00:00    frostbyte_neo

13
5

017adde4e9 · perf(amd): reuse sigmoid routing kernels across row counts · 最后更新于 2026-09-10 16:59:10 +00:00    frostbyte_neo

13
3

e836cc5ce3 · merge: update Iris MoE policy stack · 最后更新于 2026-09-10 16:56:13 +00:00    frostbyte_neo

13
14

616c720b5c · merge: resolve Iris preallocation conflicts with main · 最后更新于 2026-09-10 16:55:29 +00:00    frostbyte_neo

13
12

d8929cd218 · perf(gptoss): add gfx1250 softmax top-k routing · 最后更新于 2026-09-10 16:36:53 +00:00    frostbyte_neo

17
1

be4da55446 · fix(comm): preserve zero producer-direct capacity · 最后更新于 2026-09-10 01:13:54 +00:00    frostbyte_neo

26
7

84478e96d2 · test(kda): vectorize MTP reference batch · 最后更新于 2026-09-10 00:27:26 +00:00    frostbyte_neo

32
2

f8f6f2b54a · fix(kimi3): tolerate missing gfx1250 AttnRes and snapshot graph replay · 最后更新于 2026-09-09 21:25:26 +00:00    frostbyte_neo

35
7

e9485f4c81 · fix(kimi3): keep auto shared-SiTU on gfx1250 large-M · 最后更新于 2026-09-09 21:25:25 +00:00    frostbyte_neo

35
4

8e8fffde5b · fix(moe): keep mixed-dtype gfx950 sigmoid top-k eligible · 最后更新于 2026-09-09 21:13:28 +00:00    frostbyte_neo

35
3

9e827e828a · fix(ci): update MI450 sim node id for projected MLA decode · 最后更新于 2026-09-09 21:05:06 +00:00    frostbyte_neo

35
3

ff22b1ade8 · style: format gfx1250 latent-input decode · 最后更新于 2026-09-09 20:54:07 +00:00    frostbyte_neo

35
2