--- license: apache-2.0 --- # EditScore-code Mirror of the EditScore source code: the `editscore` Python package, evaluation scripts and training examples for image-edit reward models. ## What it is for The repo copies the EditScore code repo. It holds the `editscore` package, which loads an EditScore reward model and scores an image edit against its instruction. It also holds evaluation scripts that benchmark reward models on EditReward-Bench, and examples for training EditScore reward models, best-of-N selection and reinforcement fine-tuning of image editing models. The model weights live in separate repos. ## Provenance - Upstream source: [VectorSpaceLab/EditScore](https://github.com/VectorSpaceLab/EditScore). Paper: [arXiv 2509.23909](https://arxiv.org/abs/2509.23909). Project page: [EditScore](https://vectorspacelab.github.io/EditScore). - Related upstream data: [EditReward-Bench](https://huggingface.co/datasets/EditScore/EditReward-Bench), [EditScore-Reward-Data](https://huggingface.co/datasets/EditScore/EditScore-Reward-Data), [EditScore-RL-Data](https://huggingface.co/datasets/EditScore/EditScore-RL-Data). Models: the [EditScore collection](https://huggingface.co/collections/EditScore/editscore-68d8e27ee676981221db3cfe). ## Use ```sh hf download chibifire/EditScore-code --local-dir EditScore cd EditScore pip install -e . bash evaluate.sh ``` Install PyTorch for your CUDA version first. `vllm` is optional and speeds up inference. See the [upstream README](https://github.com/VectorSpaceLab/EditScore) for full setup, usage and the training and RL guides under `examples/`. ## Licence Apache 2.0, the licence of the [EditScore code repo](https://github.com/VectorSpaceLab/EditScore). Terms: [LICENSE](https://github.com/VectorSpaceLab/EditScore/blob/main/LICENSE).