Record the reference-free text-accuracy sweep

README.md

@@ -50,6 +50,10 @@ Only the transformer is quantized. The Qwen3-VL text encoder stays bf16 and is a
 16.3 GB of a 21.5 GB resident total, so it, not this file, decides how many reference
 images fit on a card.
 
+## Text accuracy against bf16
+
+The nvfp4 build of this same recipe scored 15/16 against bf16's 14/16 on a reference-free seed sweep (8 seeds, two text prompts, text read against the prompt) — indistinguishable. This int4 build was not put through that sweep; its LPIPS is 0.1836 against the nvfp4 build's 0.1425, so treat it as the same recipe at slightly lower fidelity, on the far wider set of cards it runs on.
+
 ## Requirements
 
 Packed for a specific kernel. The layout is an MMA fragment swizzle, so a different