Upload converted model from /home/runner/work/layaForWeb/layaForWeb/build/model for ONNX Runtime Web
cd232d2 verified |
Download README.md from VishalMysore/layaForWeb: direct link, hf CLI and curl.
- Browser
- Download file 1.02 kB
-
https://huggingface.co/VishalMysore/layaForWeb/resolve/main/README.md
- Command line
-
hf download hf://VishalMysore/layaForWeb/README.md
-
curl -L -o README.md https://huggingface.co/VishalMysore/layaForWeb/resolve/main/README.md
1.02 kB
| license: apache-2.0 | |
| base_model: convaiinnovations/laya | |
| library_name: onnx | |
| tags: [onnx, onnxruntime-web, browser, decision-model, quantized] | |
| # laya converted for the browser | |
| This is a modified copy of the checkpoint at [convaiinnovations/laya](https://huggingface.co/convaiinnovations/laya) | |
| (Apache-2.0, Copyright ConvAI Innovations), made for ONNX Runtime Web. It is unofficial and not affiliated with | |
| ConvAI Innovations. This checkpoint is a general-purpose checkpoint. Laya itself is built on ModernBERT-large by Answer.AI and LightOn | |
| (Apache-2.0). | |
| **Changes from the original:** exported to ONNX; weights quantized (weight-only int8 or int4, int8 embeddings), so | |
| outputs differ slightly from the original PyTorch model; weight files split into parts (`*.onnx.data.partNNN`, listed | |
| in `manifest.json`, which the demo page reassembles). | |
| Licensed under the Apache License, Version 2.0. See `LICENSE` and `NOTICE.md` in this folder. | |
| Build scripts and demo: the `layaForWeb` repository that produced this folder. | |