Finalize public CUDA control documentation

README.md

@@ -13,14 +13,22 @@ license: apache-2.0
 # Embedding Quantization CUDA Control
 
 A bounded ZeroGPU control lane for the matched local MLX embedding
-quantization experiment. The first milestone reproduces the frozen
-Qwen3-Embedding-0.6B BF16 vectors on CUDA and compares them with the saved MLX
-BF16 vectors. CUDA results do not reproduce MLX/Metal performance.
+quantization experiment. It reproduces the frozen Qwen3-Embedding-0.6B BF16
+vectors on CUDA and compares them with the saved MLX BF16 vectors. CUDA results
+do not reproduce MLX/Metal performance.
 
 The second milestone adds CUDA-native bitsandbytes INT8 and NF4 controls.
 They are deliberately reported as separate quantizers and are not treated as
 equivalents of MLX Q, oQ, or oQe formats.
 
-Matched BF16 controls are also available for GTE-Qwen2-1.5B and
-Qwen3-Embedding-8B. Their CUDA vectors are compared against the corresponding
-saved MLX BF16 vectors from the private reproduction bucket.
+Matched BF16, INT8, and NF4 controls are available for
+Qwen3-Embedding-0.6B, GTE-Qwen2-1.5B, and Qwen3-Embedding-8B. Quantized CUDA
+vectors are evaluated against their same-loading-path CUDA BF16 reference.
+
+GTE-Qwen2-1.5B failed the loading-path and MLX/CUDA parity checks. Its CUDA
+quantized results remain valid only as same-path comparisons against the direct
+CUDA BF16 reference and must not be mixed into cross-runtime vector trends.
+
+Frozen inputs, saved vectors, raw metrics, locked model revisions, and the
+portable client are published separately in
+[`TiGa-RCE/embedding-quant-repro-2026-07-28`](https://huggingface.co/datasets/TiGa-RCE/embedding-quant-repro-2026-07-28).