EmbeddingGemma 2 · ONNX int8 — FaceNest mirror (Windows / Linux)
What this is. A copy of the model files FaceNest loads for semantic search (text→image / image→image), mirrored so a given app build always gets the weights it was measured against. It is a subset of the upstream repository — only the files the app loads, nothing else.
| Upstream repository | onnx-community/embeddinggemma-2-ONNX |
| Upstream revision | daa72c51243991dfcaf9f9137d2c573d8f7790c0 |
| Model base | google/embeddinggemma-2 |
| License | Apache-2.0 |
| Files | 11 (541,793,329 bytes) |
Every weight file below is an unmodified copy of the upstream object: its SHA-256 equals the SHA-256 recorded in the upstream repository's LFS metadata (verified 2026-10-15).
Files
| file | bytes | sha256 |
|---|---|---|
chat_template.jinja |
1,016 | 4b852efc0b9960283e735363331e6f325b33bc74bdbaa076f595bc4e9b94d85e |
config.json |
5,031 | 8d011bfe08b5e345bbe0b81e5c6fd02c381920b345b986047bc2a33ce7b90d1d |
config_sentence_transformers.json |
1,565 | 031e56a498d33c349ab489a21885bcfe25b4fcba841149dc99e1e90d4a7c28f5 |
onnx/model_quantized.onnx |
495,165 | d06edd601f851c633a2519304cbeb8dc6170d7ceb61b436625c17fb9b6e74953 |
onnx/model_quantized.onnx_data |
313,724,928 | 278a7ff1248c3618e4bd11a607fc54f7bdc7778854230f3956d3f86bd9db4f3b |
onnx/vision_encoder_quantized.onnx |
162,495 | bb0de2df53a2448a32dc7908a187c168c8afd514d4d6f674f7f46024875fa4e3 |
onnx/vision_encoder_quantized.onnx_data |
195,228,672 | 3dabd69c0a36e9a8771ad82030dde74daa5a0e02b7047a5d3f3382b1137bab89 |
preprocessor_config.json |
560 | 9344893f8d0573a46ebb2ca54c03f56d044cea194c8dcfb1f4d652240ac21daa |
processor_config.json |
1,788 | 168f6a08522f3ce5dea596d94d003af2fd691742d4f41fe1f9d8cce76bfbf69c |
tokenizer.json |
32,170,510 | 4d777ef5bdc1aa36227abdfb77c3e49e7b9c892d16e1b6bda41c393504828be4 |
tokenizer_config.json |
1,599 | 17bd5d6e9364ca49a534e1502076593317c298d4a663623091ed45388f004874 |
Why a mirror
The upstream ONNX repository is community-maintained: it can be re-quantized, re-uploaded or removed without notice. FaceNest pins a revision and serves a stable copy, so an app build never silently gets different weights than the ones its thresholds were calibrated against.
Note on the ONNX layout
onnx/model_quantized.onnx and onnx/vision_encoder_quantized.onnx are the
graphs; the tensor weights live in the sibling *_data files (ONNX external
data, the standard large-model layout). Both must be present side by side —
the .onnx file on its own is not loadable.
License
Apache-2.0, as declared by the upstream repositories. This mirror adds no additional terms.
- Downloads last month
- 18