Upload h/mxfp8/README.md with huggingface_hub
Browse files- h/mxfp8/README.md +1 -1
h/mxfp8/README.md
CHANGED
|
@@ -52,7 +52,7 @@ This is lower precision than the 8-bit affine bundle. Treat this as experimental
|
|
| 52 |
|
| 53 |
## Measured Speed
|
| 54 |
|
| 55 |
-
Packed
|
| 56 |
|
| 57 |
| p50 latency | Throughput |
|
| 58 |
| ---: | ---: |
|
|
|
|
| 52 |
|
| 53 |
## Measured Speed
|
| 54 |
|
| 55 |
+
Packed weight-only runtime, fast-kernel compiled-forward MLX at `512x512`, batch 1:
|
| 56 |
|
| 57 |
| p50 latency | Throughput |
|
| 58 |
| ---: | ---: |
|