 
|
b501311fa1
|
[Kimi K3] Fuse MLA gate projection into QKV-A GEMM (#33623)
Co-authored-by: Xiaoyu Zhang <1182563586@qq.com>
Co-authored-by: Brayden Zhong <b8zhong@uwaterloo.ca>
|
2026-08-13 00:21:34 +08:00 |
|
 Chenchen HongandXiaoyu Zhang
|
49b9c46f41
|
[dLLM] Reuse block KV/req slots in place across FDFO rounds (#27877)
Co-authored-by: Xiaoyu Zhang <1182563586@qq.com>
|
2026-07-20 13:36:13 +08:00 |
|
 Chenchen HongandXiaoyu Zhang
|
e3ceccf781
|
[dLLM] Make FDFO a framework capability for all dLLM algorithms (#27551)
Co-authored-by: Xiaoyu Zhang <1182563586@qq.com>
|
2026-07-11 11:05:05 +08:00 |
|