Document measured H3ddle performance

README.md

@@ -47,16 +47,24 @@ The resulting checkpoint is **not compatible with runtimes that assume the
 original output-major Comfy layout**. That original remains available from
 [Comfy-Org/MiniMax-H3](https://huggingface.co/Comfy-Org/MiniMax-H3).
 
-## Validation
+## Measured performance versus the regular layout
 
 The complete artifact was checked against the source tensor by tensor: all 200
 converted matrices matched the exact transpose, all 732 unchanged tensors were
 byte-identical, and the native 50-layer output-major/input-major Metal smoke
 tests produced identical video and audio hashes.
 
-On a 32 GiB M1 Pro, the same input-major kernel and format reduced a matched
-512x512 Turbo generation's transformer time by 8.9%. The standard checkpoint
-in this repository also passed a real H3ddle image-quality check before
+The directly timed A/B used the same layout and kernel on a matched 512x512
+Turbo generation on a 32 GiB M1 Pro; it is representative of the projection
+path, not a separately timed standard-model claim.
+
+| path | transformer | total generation | gain vs regular |
+|---|---:|---:|---:|
+| regular output-major | 253.9 s | 278.7 s | baseline |
+| input-major | 231.4 s | 256.5 s | **8.9% transformer; 8.0% total (1.09x)** |
+
+The generated images were identical. The standard checkpoint in this
+repository separately passed a real H3ddle image-quality check before
 publication. Actual improvement varies with canvas size and storage pressure.
 
 ## H3ddle package layout