Commit Graph
6 Commits
Author SHA1 Message Date
946a6fddd5 Fix MoT RMSNorm for vLLM 0.21
Use the vLLM IR RMSNorm op after the legacy layernorm helper was removed, keeping MoT fused paths unchanged while restoring BAGEL compatibility.

Co-authored-by: Yufeng Zhu <yzhu802@gatech.edu>
Co-authored-by: Cursor <cursoragent@cursor.com>
Signed-off-by: princepride <wangzhipeng628@gmail.com>
2026-05-29 02:34:52 +00:00
Yufeng Zhu 424fa3fcfc Address review comments (benchmark logs, ids, and design comments)
Signed-off-by: Yufeng Zhu <yzhu802@gatech.edu>
Signed-off-by: princepride <wangzhipeng628@gmail.com>
2026-05-29 02:34:52 +00:00
Zhengyuan SuandClaude Opus 4.6 e1834ab19b [Enhancement] Fix SP path for MoT fused kernels, address review feedback
Rebase MoT fused kernels (PR #1328) onto current main and fix issues:

- Rewrite _forward_sp_gen() to use MoT unified API instead of deleted
  *_moe_gen layers (qkv_proj_moe_gen, o_proj_moe_gen, q_norm_moe_gen,
  k_norm_moe_gen), which caused AttributeError when SP was active
- Parameterize bias in MoT kernel tests (ZJY0516 review feedback)
- Add configs/ directory with README documenting the 3-tier config
  loading mechanism (ZJY0516 review feedback)

Signed-off-by: Zhengyuan Su <su.zhengyuan@u.nus.edu>
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Signed-off-by: princepride <wangzhipeng628@gmail.com>
2026-05-29 02:34:52 +00:00
汪志鹏 b52da5455f Add pytest markers for core model and diffusion tests
Signed-off-by: 汪志鹏 <wangzhipeng628@gmail.com>
Signed-off-by: princepride <wangzhipeng628@gmail.com>
2026-05-29 02:34:52 +00:00
汪志鹏 5958b01ff7 Add pytest markers for core model and diffusion tests
Signed-off-by: 汪志鹏 <wangzhipeng628@gmail.com>
Signed-off-by: princepride <wangzhipeng628@gmail.com>
2026-05-29 02:34:51 +00:00
Yufeng Zhu f9ba582747 enhancement: add mot fused kernels and benchmarks 01
Signed-off-by: Yufeng Zhu <yzhu802@gatech.edu>
Signed-off-by: princepride <wangzhipeng628@gmail.com>
2026-05-29 02:34:51 +00:00