Upload README.md with huggingface_hub

README.md

@@ -31,14 +31,17 @@ Getting it wrong does not throw; it returns vectors that look fine and rank wron
 
 ## Verification
 
-| build | file | size (MB) | Mac ms* | worst cosine vs eager | retrieval budget |
-|---|---|---|---|---|---|
-| fp32 | `embed_all_mpnet_xnnpack_fp32.pte` | 435.8 | 35.2 | 1.000000 | 0% |
-| fp16 | `embed_all_mpnet_xnnpack_fp16.pte` | 218.1 | 56.2 | 1.000000 | 11% |
-| Core ML (fp16, iOS) | `embed_all_mpnet_coreml_all.pte` | 220.2 | 6.2 | 0.999993 | 32% |
+| build | file | size (MB) | Mac ms* | backend takes | worst cosine vs eager | retrieval budget |
+|---|---|---|---|---|---|---|
+| fp32 | `embed_all_mpnet_xnnpack_fp32.pte` | 435.8 | 35.2 | 75.5% | 1.000000 | 0% |
+| fp16 | `embed_all_mpnet_xnnpack_fp16.pte` | 218.1 | 56.2 | 65.7% | 1.000000 | 11% |
+| Core ML (fp16, iOS) | `embed_all_mpnet_coreml_all.pte` | 220.2 | 6.2 | 100.0% | 0.999993 | 32% |
 
-\*Mac arm64, median of 10, one 256-token sequence — a reference point for relative
-cost, not a device number. Torch eager fp32 on the same machine is
+\*Mac arm64, one 256-token sequence, **fastest of five medians of ten** — a reference
+point for relative cost, not a device number. The host shares its cores with other work,
+and a single median does not survive that: the same eager model here measured 19.6 ms and
+182.8 ms twenty minutes apart. Contention only ever adds time, so the fastest repetition is
+the one that means something. Torch eager fp32, measured the same way, is
 37.0 ms.
 
 Cosine is measured against the model run in eager through its own pooling, over eight