biondizzle/nvfp4-megamoe-kernel
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Update STATUS.md: E5 done

Browse Source
This commit is contained in:
biondizzle
2026-05-30 21:21:47 +00:00
parent df6220abaf
commit afc07a5d1a
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
STATUS.md
View File

@@ -27,7 +27,7 @@
- [x] **E2:** End-to-end smoke test through one full layer ✅ (SWA + CSA + HCA)
- [x] **E3:** Top-level `model/dsv4.py` ✅
- [x] **E4:** Delete `torch.cuda.synchronize()` from fast path ✅
- [ ] **E5:** Fold batch loop into kernel grid
- [x] **E5:** Fold batch loop into kernel grid ✅ (single launch for batched decode)
- [ ] **E6:** FP4 output fusion for FMHA → wo_a
- [ ] **E7:** Lightning indexer FP4 tensor-core scoring
- [ ] **E8:** Multi-CTA grid for prefill