{"slug": "ml-sharp", "name": "ml-sharp", "category": "media", "category_name": "Media and creative", "description": "", "recommenders": 4, "commenters": 10, "warners": 2, "score": 2.014, "rising": -1.31, "first": "2025-12-27", "latest": "2026-05-19", "why": "It falls apart once you move too much, but for a little side-wiggling or a second-eye view for VR, it's great.", "why_is_quote": true, "has_verdict": false, "sentiment": {"recommend": 4, "neutral": 3, "compare": 1, "warn": 2}, "spark": [0, 1, 2, 0, 0, 0, 1, 0, 0, 0, 0, 0], "listed": true, "loved": [], "caveat": "", "short_quote": "It falls apart once you move too much, but for a little side-wiggling or a second-eye view for VR, it's great.", "kind": "", "icon": true, "domain": null, "github_repo": "github.com/apple/ml-sharp", "official_url": "https://github.com/apple/ml-sharp", "pricing": "", "aliases": ["ml-sharp"], "verdict": null, "praise": [{"label": "", "summary": "", "quotes": [{"id": 771100624, "item_id": 48193789, "url": "https://news.ycombinator.com/item?id=48193789", "by": "sovok", "date": "2026-05-19", "version": null, "stance": "recommend", "quote": "It falls apart once you move too much, but for a little side-wiggling or a second-eye view for VR, it's great.", "excerpt": "Also interesting: https://github.com/apple/ml-sharp Apples model to generate Gaussian splats from a single image. Takes about 30 seconds on an M1 Pro. It falls apart once you move too much, but for a little side-wiggling or a second-eye view for VR, it's great. And looks a lot better than the old approach of depth map + vertex shaders that I use in https://github.com/combatwombat/tiefling. But ml-sharp has 2.6 GB wei", "context": "", "use_case": "Gaussian splats from one image", "switched_from": "", "story": "Gaussian Splat of a Strawberry"}, {"id": 744976832, "item_id": 46561052, "url": "https://news.ycombinator.com/item?id=46561052", "by": "james2doyle", "date": "2026-01-09", "version": null, "stance": "recommend", "quote": "I think this one is better: https://huggingface.co/spaces/ronedgecomb/ml-sharp", "excerpt": "I think this one is better: https://huggingface.co/spaces/ronedgecomb/ml-sharp", "context": "", "use_case": "3D Gaussian splat generation", "switched_from": "", "story": "Revelium Studio turns a single image into a navigable 3D Gaussian Splat model"}, {"id": 744973985, "item_id": 46560874, "url": "https://news.ycombinator.com/item?id=46560874", "by": "vunderba", "date": "2026-01-09", "version": null, "stance": "recommend", "quote": "It's pretty trivial to get running locally and generating the PLY files.", "excerpt": "Yeah I think you're right. It calls that out (in really tiny footer text) that it's leveraging ml-sharp. It's pretty trivial to get running locally and generating the PLY files. Spark's a pretty good renderer for it after you've generated the gaussian splats. https://github.com/sparkjsdev/spark", "context": "", "use_case": "", "switched_from": "", "story": "Revelium Studio turns a single image into a navigable 3D Gaussian Splat model"}, {"id": 742432704, "item_id": 46402044, "url": "https://news.ycombinator.com/item?id=46402044", "by": "neom", "date": "2025-12-27", "version": null, "stance": "recommend", "quote": "Examples: https://apple.github.io/ml-sharp/", "excerpt": "Examples: https://apple.github.io/ml-sharp/ Paper: https://arxiv.org/abs/2512.10685", "context": "", "use_case": "", "switched_from": "", "story": "Apple releases open-source model that instantly turns 2D photos into 3D views"}]}], "caveats": [{"label": "", "summary": "", "quotes": [{"id": 742449584, "item_id": 46403099, "url": "https://news.ycombinator.com/item?id=46403099", "by": "runjake", "date": "2025-12-27", "version": null, "stance": "warn", "quote": "it seems like one is stuck with using this for personal use?", "excerpt": "Link to the actual project license, since it hasn't been referenced yet: https://github.com/apple/ml-sharp/blob/main/LICENSE Between this and the model's license, it seems like one is stuck with using this for personal use?", "context": "", "use_case": "", "switched_from": "", "story": "Apple releases open-source model that instantly turns 2D photos into 3D views"}, {"id": 742434864, "item_id": 46402179, "url": "https://news.ycombinator.com/item?id=46402179", "by": "RobotToaster", "date": "2025-12-27", "version": null, "stance": "warn", "quote": "\"Exclusively for research purposes\" so not actually open source.", "excerpt": "https://raw.githubusercontent.com/apple/ml-sharp/refs/heads/main/LICENSE_MODEL \"Exclusively for research purposes\" so not actually open source.", "context": "", "use_case": "", "switched_from": "", "story": "Apple releases open-source model that instantly turns 2D photos into 3D views"}]}], "who": [], "tags": [], "use_cases": ["gaussian splat generation", "gaussian splats from one image", "3d gaussian splat generation", "photo to 3d gaussian splat", "3d views from photos", "single-image view synthesis"], "compared": [], "makers": [{"id": 742449408, "item_id": 46403088, "url": "https://news.ycombinator.com/item?id=46403088", "by": "rcarmo", "date": "2025-12-27", "version": null, "stance": "neutral", "quote": "A couple of weeks ago I forked it and got the rendering to work in MPS: https://github.com/rcarmo/ml-sharp", "excerpt": "This is a dupe. A couple of weeks ago I forked it and got the rendering to work in MPS: https://github.com/rcarmo/ml-sharp", "context": "", "use_case": "", "switched_from": "", "story": "Apple releases open-source model that instantly turns 2D photos into 3D views"}, {"id": 740620944, "item_id": 46288809, "url": "https://news.ycombinator.com/item?id=46288809", "by": "rcarmo", "date": "2025-12-16", "version": null, "stance": "recommend", "quote": "Fixed that: https://github.com/rcarmo/ml-sharp", "excerpt": "Fixed that: https://github.com/rcarmo/ml-sharp", "context": "", "use_case": "Single-image view synthesis", "switched_from": "", "story": "SHARP, an approach to photorealistic view synthesis from a single image"}], "threads": [{"url": "https://news.ycombinator.com/item?id=46401539", "title": "Apple releases open-source model that instantly turns 2D photos into 3D views", "mentions": 6}, {"url": "https://news.ycombinator.com/item?id=46557352", "title": "Revelium Studio turns a single image into a navigable 3D Gaussian Splat model", "mentions": 3}, {"url": "https://news.ycombinator.com/item?id=48191602", "title": "Gaussian Splat of a Strawberry", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=46284658", "title": "SHARP, an approach to photorealistic view synthesis from a single image", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=46420401", "title": "Web Viewer for Apple's ml-sharp with 3D Gaussian Splat Rendering in the Browser", "mentions": 1}], "sentiment_by_month": {"2025-11": {}, "2025-12": {"recommend": 1, "warn": 2, "neutral": 2, "compare": 1}, "2026-01": {"neutral": 1, "recommend": 2}, "2026-02": {}, "2026-03": {}, "2026-04": {}, "2026-05": {"recommend": 1}, "2026-06": {}, "2026-07": {}, "2026-08": {}, "2026-09": {}, "2026-10": {}}, "versions": [], "version_unnamed": 0, "family_ref": null, "runs_on": null, "lines": [], "quotes": [{"id": 771103472, "item_id": 48193967, "url": "https://news.ycombinator.com/item?id=48193967", "by": "sovok", "date": "2026-05-19", "version": null, "stance": "neutral", "quote": "The Python process peaks at about 15 GB", "excerpt": "The Python process peaks at about 15 GB, so maybe a 24 GB GPU works. But others seem to have problems wih Nvidia: https://github.com/apple/ml-sharp/issues/84", "context": "", "use_case": "Gaussian splat generation", "switched_from": "", "story": "Gaussian Splat of a Strawberry"}, {"id": 771100624, "item_id": 48193789, "url": "https://news.ycombinator.com/item?id=48193789", "by": "sovok", "date": "2026-05-19", "version": null, "stance": "recommend", "quote": "It falls apart once you move too much, but for a little side-wiggling or a second-eye view for VR, it's great.", "excerpt": "Also interesting: https://github.com/apple/ml-sharp Apples model to generate Gaussian splats from a single image. Takes about 30 seconds on an M1 Pro. It falls apart once you move too much, but for a little side-wiggling or a second-eye view for VR, it's great. And looks a lot better than the old approach of depth map + vertex shaders that I use in https://github.com/combatwombat/tiefling. But ml-sharp has 2.6 GB wei", "context": "", "use_case": "Gaussian splats from one image", "switched_from": "", "story": "Gaussian Splat of a Strawberry"}, {"id": 744976832, "item_id": 46561052, "url": "https://news.ycombinator.com/item?id=46561052", "by": "james2doyle", "date": "2026-01-09", "version": null, "stance": "recommend", "quote": "I think this one is better: https://huggingface.co/spaces/ronedgecomb/ml-sharp", "excerpt": "I think this one is better: https://huggingface.co/spaces/ronedgecomb/ml-sharp", "context": "", "use_case": "3D Gaussian splat generation", "switched_from": "", "story": "Revelium Studio turns a single image into a navigable 3D Gaussian Splat model"}, {"id": 744973985, "item_id": 46560874, "url": "https://news.ycombinator.com/item?id=46560874", "by": "vunderba", "date": "2026-01-09", "version": null, "stance": "recommend", "quote": "It's pretty trivial to get running locally and generating the PLY files.", "excerpt": "Yeah I think you're right. It calls that out (in really tiny footer text) that it's leveraging ml-sharp. It's pretty trivial to get running locally and generating the PLY files. Spark's a pretty good renderer for it after you've generated the gaussian splats. https://github.com/sparkjsdev/spark", "context": "", "use_case": "", "switched_from": "", "story": "Revelium Studio turns a single image into a navigable 3D Gaussian Splat model"}, {"id": 744966304, "item_id": 46560394, "url": "https://news.ycombinator.com/item?id=46560394", "by": "personjerry", "date": "2026-01-09", "version": null, "stance": "neutral", "quote": "This is the heavy lifting: https://github.com/apple/ml-sharp", "excerpt": "This is just Apple's tool plus a splat viewing library? Perhaps disingenuous to call \"our web app\" This is the heavy lifting: https://github.com/apple/ml-sharp Previous discussion: https://news.ycombinator.com/item?id=46284658", "context": "", "use_case": "", "switched_from": "", "story": "Revelium Studio turns a single image into a navigable 3D Gaussian Splat model"}, {"id": 742726432, "item_id": 46420402, "url": "https://news.ycombinator.com/item?id=46420402", "by": "kstonekuan", "date": "2025-12-29", "version": null, "stance": "neutral", "quote": "Gaussian Splat .ply files generated by Apple's ml-sharp", "excerpt": "I built a web viewer for visualizing Gaussian Splat .ply files generated by Apple's ml-sharp, which converts a single photo into a 3D Gaussian Splat in under a second [1]. Features: - Upload and view .ply file directly in the browser - Multiple camera trajectory animations (rotate, swipe, shake, forward) - Interactive orbit controls (drag to orbit, scroll to zoom, right-drag to pan) - No installation required \u2013 runs ", "context": "", "use_case": "photo to 3D Gaussian splat", "switched_from": "", "story": "Web Viewer for Apple's ml-sharp with 3D Gaussian Splat Rendering in the Browser"}, {"id": 742454610, "item_id": 46403413, "url": "https://news.ycombinator.com/item?id=46403413", "by": "avaer", "date": "2025-12-27", "version": null, "stance": "compare", "quote": "although HWM itself does support things like multi-image merging (which ml-sharp does not)", "excerpt": "It doesn't but it's pretty trivial to do if all you want is a pinholed mesh. I managed to one-shot it by mixing in the mesh exporter from https://github.com/Tencent-Hunyuan/HunyuanWorld-Mirror but at that point you might as well use HWM, which is slower but much better suited to the level design use case. Note that the results might not be as good as you expect, because this does not do any angled inpainting -- any d", "context": "", "use_case": "3D views from photos", "switched_from": "", "story": "Apple releases open-source model that instantly turns 2D photos into 3D views"}, {"id": 742450784, "item_id": 46403174, "url": "https://news.ycombinator.com/item?id=46403174", "by": "jonas21", "date": "2025-12-27", "version": null, "stance": "neutral", "quote": "The weights are under a different, more restrictive, license.", "excerpt": "The code is licensed [1] under the \"Apple MIT\" license [2], which is considered open-source. The weights are under a different, more restrictive, license. This is mentioned at the bottom of the README. [1] https://github.com/apple/ml-sharp/blob/main/LICENSE [2] https://fedoraproject.org/wiki/Licensing/Apple_MIT_License", "context": "", "use_case": "", "switched_from": "", "story": "Apple releases open-source model that instantly turns 2D photos into 3D views"}, {"id": 742449584, "item_id": 46403099, "url": "https://news.ycombinator.com/item?id=46403099", "by": "runjake", "date": "2025-12-27", "version": null, "stance": "warn", "quote": "it seems like one is stuck with using this for personal use?", "excerpt": "Link to the actual project license, since it hasn't been referenced yet: https://github.com/apple/ml-sharp/blob/main/LICENSE Between this and the model's license, it seems like one is stuck with using this for personal use?", "context": "", "use_case": "", "switched_from": "", "story": "Apple releases open-source model that instantly turns 2D photos into 3D views"}, {"id": 742434864, "item_id": 46402179, "url": "https://news.ycombinator.com/item?id=46402179", "by": "RobotToaster", "date": "2025-12-27", "version": null, "stance": "warn", "quote": "\"Exclusively for research purposes\" so not actually open source.", "excerpt": "https://raw.githubusercontent.com/apple/ml-sharp/refs/heads/main/LICENSE_MODEL \"Exclusively for research purposes\" so not actually open source.", "context": "", "use_case": "", "switched_from": "", "story": "Apple releases open-source model that instantly turns 2D photos into 3D views"}, {"id": 742432704, "item_id": 46402044, "url": "https://news.ycombinator.com/item?id=46402044", "by": "neom", "date": "2025-12-27", "version": null, "stance": "recommend", "quote": "Examples: https://apple.github.io/ml-sharp/", "excerpt": "Examples: https://apple.github.io/ml-sharp/ Paper: https://arxiv.org/abs/2512.10685", "context": "", "use_case": "", "switched_from": "", "story": "Apple releases open-source model that instantly turns 2D photos into 3D views"}, {"id": 740603136, "item_id": 46287696, "url": "https://news.ycombinator.com/item?id=46287696", "by": "rcarmo", "date": "2025-12-16", "version": null, "stance": "neutral", "quote": "Well, I got _something_ to work on Apple Silicon", "excerpt": "Well, I got _something_ to work on Apple Silicon: https://github.com/rcarmo/ml-sharp (has a little demo GIF) I am looking at ways to approximate Gaussian splats without having to reinvent the wheel, but I'm a bit over my depth since I haven't been playing a lot of attention to those in general.", "context": "", "use_case": "single-image view synthesis", "switched_from": "", "story": "SHARP, an approach to photorealistic view synthesis from a single image"}]}