OXR-1.0, compiled for the browser
SoMark's OXR-1.0 compiled into a fusor-ai bundle: the weights the browser runtime of oxr-wasm loads to read documents to Markdown on the device's CPU. Try it in the Space fusorlabs/oxr-wasm.
- Source:
SoMarkAI/OXR-1.0at revision520bed7e90f8f4ffa110db47bdb64d5cf80ccfba, compiled with themixprofile: 4-bit visual head, everything else 7-bit (1420 MB). - Format:
manifest.jsonlists every tensor and part; each part is named by its BLAKE3 hash, which the runtime checks as it loads. - Reproduce it:
just modelin oxr-wasm downloads the checkpoint and compiles the same bundle.
This is not a PyTorch or safetensors checkpoint: it only loads in the oxr-wasm runtime.
The model is SoMark's, under the OXR License (Apache-2.0 with additional terms), which applies to this bundle; a service built on it must credit SoMark OXR.
Model tree for fusorlabs/oxr-1.0-web
Base model
SoMarkAI/OXR-1.0