docs: record GROOT N1.7 Thor artifact validation
Browse files- benchmarks/RESULTS.md +9 -0
benchmarks/RESULTS.md
CHANGED
|
@@ -59,3 +59,12 @@ were bit-exact in every row.
|
|
| 59 |
|
| 60 |
The table reports the uploaded SM87+SM110a fat binary on Thor. SM87 is present
|
| 61 |
as native cubin code but is not included in these runtime measurements.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 59 |
|
| 60 |
The table reports the uploaded SM87+SM110a fat binary on Thor. SM87 is present
|
| 61 |
as native cubin code but is not included in these runtime measurements.
|
| 62 |
+
|
| 63 |
+
### GROOT N1.7 FP4 producer update
|
| 64 |
+
|
| 65 |
+
The additive FP4 producer from FlashRT
|
| 66 |
+
`24df793f4fa2d50780aea03b644208c6e0cb4162` was rebuilt as
|
| 67 |
+
`torch213-cxx11-cu130-aarch64-linux` on the same Thor host. Packed FP4 values
|
| 68 |
+
and SFA bytes were exact against the native staged contract. The Tensor
|
| 69 |
+
wrapper/raw registered-op measurement was `8.3094/8.1932 us` (`1.0142x`) in
|
| 70 |
+
direct mode and `5.3749/5.3797 us` (`0.9991x`) under CUDA Graph.
|