{"id":49830540,"url":"https://github.com/col14m/cadrille","last_synced_at":"2026-05-13T20:38:16.974Z","repository":{"id":296903455,"uuid":"991366456","full_name":"col14m/cadrille","owner":"col14m","description":"[ICLR2026] cadrille: Multi-modal CAD Reconstruction with Online Reinforcement Learning","archived":false,"fork":false,"pushed_at":"2026-02-01T16:04:35.000Z","size":533,"stargazers_count":94,"open_issues_count":6,"forks_count":13,"subscribers_count":5,"default_branch":"master","last_synced_at":"2026-02-02T01:14:13.507Z","etag":null,"topics":["cad","cadquery","llm","pytorch","qwen2-vl","vlm"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/col14m.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2025-05-27T14:12:55.000Z","updated_at":"2026-02-01T19:17:13.000Z","dependencies_parsed_at":"2025-06-03T07:48:40.725Z","dependency_job_id":"8c29a1c3-6011-4ced-a643-5c98644053e3","html_url":"https://github.com/col14m/cadrille","commit_stats":null,"previous_names":["col14m/cadrille"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/col14m/cadrille","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/col14m%2Fcadrille","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/col14m%2Fcadrille/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/col14m%2Fcadrille/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/col14m%2Fcadrille/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/col14m","download_url":"https://codeload.github.com/col14m/cadrille/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/col14m%2Fcadrille/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":32999520,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-13T13:14:54.681Z","status":"ssl_error","status_checked_at":"2026-05-13T13:14:51.610Z","response_time":115,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["cad","cadquery","llm","pytorch","qwen2-vl","vlm"],"created_at":"2026-05-13T20:38:16.488Z","updated_at":"2026-05-13T20:38:16.963Z","avatar_url":"https://github.com/col14m.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"## `cadrille`: Multi-modal CAD Reconstruction with Online Reinforcement Learning\n\n**News**:\n * :fire: Jan, 2026. `cadrille` is accepted to ICLR 2026.\n * :fire: May, 2025. `cadrille` is state-of-the-art in three CAD reconstruction benchmarks: DeepCAD, Fusion360, CC3D.\n \nThis repository contains an implementation of `cadrille`, a multi-modal (point clouds / images / text) 3D CAD reconstruction method introduced in our paper:\n\n\u003e **cadrille: Multi-modal CAD Reconstruction with Online Reinforcement Learning**\u003cbr\u003e\n\u003e [Maksim Kolodiazhnyi](https://github.com/col14m),\n\u003e [Denis Tarasov](https://dt6a.github.io),\n\u003e [Dmitrii Zhemchuzhnikov](https://github.com/zhemdi),\n\u003e [Alexander Nikulin](https://howuhh.github.io),\n\u003e [Ilya Zisman](https://zis.mn),\n\u003e [Anna Vorontsova](https://highrut.github.io),\n\u003e [Anton Konushin](https://scholar.google.com/citations?user=ZT_k-wMAAAAJ),\n\u003e [Vladislav Kurenkov](https://dunnolab.ai),\n\u003e [Danila Rukhovich](https://github.com/filaPro) \u003cbr\u003e\n\u003e https://arxiv.org/abs/2505.22914\n\n### Installation\n\nInstall Python packages according to our [Dockerfile](Dockerfile). We support DeepCAD (test), Fusion360 (test), Text2CAD (train / val / test), and CAD-Recode (train, val) datasets. Follow our [instruction](data/README.md) to download and preprocess data.\n\n### Train\n\nTo start training run *train.py* script:\n```shell\npython train.py --mode pc_img --use-text\n```\nTo disable some of the modalities set *--mode* to *img* or *pc*, or disable *--use-text*. We don't provide RL fine-tuning code for now. Alternatively both [SFT](https://huggingface.co/maksimko123/cadrille) and [RL](https://huggingface.co/maksimko123/cadrille-rl) models can be downloaded from :hugs: HuggningFace.\n\n### Inference\n\nTo predict CadQuery codes run *test.py* script:\n```shell\npython test.py --split deepcad_test_mesh --mode pc\n```\nTo run on other datasets and modalities use *--split fusion360_test_mesh* or set *--mode* to *img* or *text*.\n\n### Evaluation\n\nTo evaluate IoU, invalidity ratio, and chamfer distance run *evaluate.py* script:\n```shell\npython evaluate.py\n```\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"https://github.com/user-attachments/assets/8b811b14-e646-48d6-9a0c-06a9655bdbaf\" alt=\"cadrille scheme\"/\u003e\n\u003c/p\u003e\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"https://github.com/user-attachments/assets/d6ae21f5-6c3c-4b7b-a2e9-ff0a310caa3d\" alt=\"cadrille predictions\"/\u003e\n\u003c/p\u003e\n\n### Citation\n\nIf you find this work useful for your research, please cite our paper:\n\n```\n@article{kolodiazhnyi2025cadrille,\n  title={cadrille: Multi-modal CAD Reconstruction with Online Reinforcement Learning},\n  author={Maksim Kolodiazhnyi, Denis Tarasov, Dmitrii Zhemchuzhnikov, Alexander Nikulin, Ilya Zisman, Anna Vorontsova, Anton Konushin, Vladislav Kurenkov, Danila Rukhovich},\n  journal={arXiv preprint arXiv:2505.22914},\n  year={2025}\n}\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcol14m%2Fcadrille","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fcol14m%2Fcadrille","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcol14m%2Fcadrille/lists"}