Add validated FP32 and INT8 ONNX exports

onnx/README.md

@@ -0,0 +1,33 @@
+# ONNX exports
+
+Source: [Shuu12121/NightJar-CodeSearch-Embedding](https://huggingface.co/Shuu12121/NightJar-CodeSearch-Embedding/tree/b9522c84bad8b17d61373e2ebd7cf53d80c406d4)
+
+| File | Precision |
+| --- | --- |
+| model.onnx | FP32 |
+| model_quantized.onnx | Dynamic INT8 (constant MatMul and Gather weights) |
+
+Inputs: `input_ids` and `attention_mask`, int64 `[batch, sequence]`.
+Output: `last_hidden_state`. Take the first token, then L2-normalize to obtain
+768-dimensional embeddings. Maximum input: 1024 tokens.
+Use the tokenizer/configuration at the repository root. Query/document prefixes are not required.
+
+Node.js short-input example (Transformers.js 4.2.0):
+
+```javascript
+import { pipeline } from '@huggingface/transformers';
+const extractor = await pipeline('feature-extraction', 'Shuu12121/NightJar-CodeSearch-Embedding', {
+  dtype: 'q8', device: 'cpu',
+});
+const output = await extractor('parse a JSON string', { pooling: 'cls', normalize: true });
+console.log(output.dims); // [1, 768]
+await extractor.dispose();
+```
+
+For long inputs, truncate the content to 1022 tokens before adding
+CLS and SEP. Transformers.js 4.2.0's default truncation slices the final sequence and
+can remove SEP; preserve both special tokens for parity with the Python tokenizer.
+
+`manifest.json` records the source commit, export settings, validation results and
+SHA256 checksums. These exports use opset 17. The original model's
+license and usage terms apply. INT8 output embeddings remain FP32.

onnx/manifest.json

@@ -0,0 +1,36 @@
+{
+  "format_version": 1,
+  "model_id": "Shuu12121/NightJar-CodeSearch-Embedding",
+  "revision": "b9522c84bad8b17d61373e2ebd7cf53d80c406d4",
+  "max_length": 1024,
+  "dimensions": 768,
+  "pooling": "cls",
+  "normalize": true,
+  "opset": 17,
+  "quantization": "dynamic QInt8, per-channel MatMul and Gather weights",
+  "files": {
+    "model.onnx": 438209760,
+    "model_quantized.onnx": 110392861
+  },
+  "validation": {
+    "model.onnx": {
+      "min_cosine_to_pytorch": 0.9999998736126565,
+      "max_abs_error": 1.7881393432617188e-07
+    },
+    "model_quantized.onnx": {
+      "min_cosine_to_pytorch": 0.9995053808395933,
+      "max_abs_error": 0.003841400146484375
+    }
+  },
+  "versions": {
+    "torch": "2.9.1",
+    "transformers": "4.57.6",
+    "sentence-transformers": "5.7.0",
+    "optimum-onnx": "0.1.0",
+    "onnxruntime": "1.29.0"
+  },
+  "sha256": {
+    "model.onnx": "80365539adaafde121c2d97a92fb0dae3c70e041dd0c2908ce53da0e3665665b",
+    "model_quantized.onnx": "4232e176e2ebc27e80e340e83c806420df3d2f5b72bf4db2f005f23ce37f7e65"
+  }
+}

onnx/model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80365539adaafde121c2d97a92fb0dae3c70e041dd0c2908ce53da0e3665665b
+size 438209760

onnx/model_quantized.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4232e176e2ebc27e80e340e83c806420df3d2f5b72bf4db2f005f23ce37f7e65
+size 110392861