cortexelus commited on
Commit
4697212
·
verified ·
1 Parent(s): a05f2d4

sa3-m fp8: ship calibrated-bakedmin engine (176 fp8 GEMM + 96 bf16 fused MHA + baked fp32 RoPE; calibration by @ryanontheinside #47)

Browse files
Files changed (1) hide show
  1. tensorRT/sm_90/sa3-m/dit_fp8.trt +2 -2
tensorRT/sm_90/sa3-m/dit_fp8.trt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2b0279a1ee9055ebf5a18e5728c867e7909fdcded9569e10475f0771f64e1d4f
3
- size 1488859956
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:88e69eab8442958352eff27fc87c504a63ef1b5aa75d7e4475edeb272b649db5
3
+ size 1492912444