--- title: OptimaMT emoji: ๐ŸŒ short_description: Lighter OPUS-MT translation models for ONNX Runtime colorFrom: blue colorTo: indigo sdk: static pinned: false --- # OptimaMT **ONNX Runtime's fast mode at half the memory: 2.3โ€“2.5ร— faster than the default, same translations.** 51 OPUS-MT translation models for ONNX Runtime and transformers.js: drop-in copies of the int8 exports with a lighter merged decoder. Same weights, same tokens. | Memory after load (median; Linux 51 pairs, Windows 49) | Linux | Windows | |---|---:|---:| | `disable_quant_qdq = 1` | 525 โ†’ 281 MiB (-46 %) | 500 โ†’ 257 MiB (-49 %) | | ONNX Runtime defaults | 277 โ†’ 217 MiB (-22 %) | 196 โ†’ 166 MiB (-16 %) | **Used in production:** - [OpenCircle](https://github.com/AmerAlreyahi/OpenCircle), a desktop screen translator for Windows and Linux, ships these models since v0.1.4. [All models](https://huggingface.co/collections/OptimaMT/optimamt-lighter-opus-mt-onnx-models-6af228dc9cd91f015e56e869) ยท repo ids `OptimaMT/-`, e.g. [`OptimaMT/en-ar`](https://huggingface.co/OptimaMT/en-ar) ยท [interactive results](https://huggingface.co/spaces/OptimaMT/results) ยท [method and code](https://github.com/AmerAlreyahi/OptimaMT) Models CC BY 4.0 (OPUS-MT, University of Helsinki; ONNX exports by Xenova); code MIT.