Record architecture-specific runtime format audit and n-gram scope

README.md

@@ -15,6 +15,17 @@ tags:
 - development-preview
 ---
 
+## Runtime format audit (2026-10-06)
+
+No quantization-container correction was needed.
+This family has no n-gram tensors; no n-gram file or declaration was added.
+This CUDA pack is outside MLX/oMLX/MTPLX export scope.
+
+See [runtime_audit.json](runtime_audit.json) for pinned config/index/header
+bindings, architecture, physical-format findings, and applied corrections.
+This is development evidence; no quality, MTP exactness, speed, or certification
+claim is added. Historical evidence stays bound to its original revision.
+
 # AX-Qwen3-Embedding-8B-CUDA-AXQ-NVFP4-W4A4
 
 **AXQuant CUDA NVFP4 W4A4 mixed precision development preview.** Converted

runtime_audit.json

@@ -0,0 +1,43 @@
+{
+  "applied_config_corrections": [],
+  "current_config_sha256": "c0c2c43724dbcc6dc6d5ffee0ecdd8b2caf787dd9dcbaee1c99077cf0b6cfe27",
+  "date": "2026-10-06",
+  "header_sha256": {
+    "model-00001-of-00004.safetensors": "32411907ffceb53ab5721f8c6f5ad77cc3a7a2c56cb3053e211f3d8a7216adac",
+    "model-00002-of-00004.safetensors": "4ccc539c50f8bfa4a554dd80dcd649292113b0e11bc03d827a5b58a408fdbb9b",
+    "model-00003-of-00004.safetensors": "6fc5f484ac650ae94f03aacfee96a21b96bcd19722fc2f96f0aefbfe2f51d83b",
+    "model-00004-of-00004.safetensors": "d9fd30277814f33e80e08bab931af73f15424f6e71ab242f27e5e1783ac19e80"
+  },
+  "input_bindings": {
+    "config.json": "c0c2c43724dbcc6dc6d5ffee0ecdd8b2caf787dd9dcbaee1c99077cf0b6cfe27",
+    "model.safetensors.index.json": "28b535b0551125ad1567b2e95b8084829497f7094329e20332dc48d2c1bdf206"
+  },
+  "issues": [],
+  "model_type": "qwen3",
+  "mtp_files": [],
+  "ngram_action": "none; do not invent n-gram data",
+  "ngram_files": [],
+  "ngram_quantization": [],
+  "ngram_table_metadata": null,
+  "ngram_tensor_count": 0,
+  "quality_certified": false,
+  "quantization": {
+    "quantization_config": {
+      "quant_method": "compressed-tensors"
+    }
+  },
+  "removed_source_evidence": [],
+  "repo_id": "AutomatosX/AX-Qwen3-Embedding-8B-CUDA-AXQ-NVFP4-W4A4",
+  "runtime_arch_id": null,
+  "runtime_verified": false,
+  "schema_version": "axquant.hub-runtime-audit.v1",
+  "scope": "Pinned remote config/index/Safetensors header audit; no runtime load or generation claim.",
+  "source_revision": "f93bb79d3b6f308750e97e129176a56622af519f",
+  "status": "outside-mlx-peer-scope",
+  "unchanged_weight_sha256": {
+    "model-00001-of-00004.safetensors": "95e7215c3735a750b6b7dba2176eab2c6b09083d9db7d87b3cc6602499d40471",
+    "model-00002-of-00004.safetensors": "5bb0706d5abe7aa61e5020dc3ccd1bca9064fd911eb1e38d1a400701f54cdadb",
+    "model-00003-of-00004.safetensors": "223ddc84244e0f70af57389fa25d41bf8fb9a4aded4a2512aeb27807862384a1",
+    "model-00004-of-00004.safetensors": "36cbc9c60375693629f25743c1e77ebb1724af58e671b2376463193c7fd21ef6"
+  }
+}