{"apiVersion":"v1","taxonomyVersion":1,"generatedAt":"2026-10-29T15:48:46.237Z","data":{"recordId":"wsa_xyrxaws4xpqw","score":1,"confidence":"high","capped":false,"capReason":null,"sources":[{"n":1,"url":"https://developers.googleblog.com/native-model-authoring-for-edge-with-litert-tensor-api/","publisher":"Google","authority":"owner-controlled","verification":"verified","lastRetrievedAt":"2026-10-28T00:15:38.667Z","detectedPublishedAt":"2026-10-19T00:00:00.000Z","rights":"original"}],"claims":[{"id":"f2b6908a-5f09-4023-b61c-d4284e187869","kind":"availability","statement":"Google announced the release of the LiteRT Tensor API.","citations":[{"source":1,"supportProbability":0.95,"supports":true,"quote":{"text":"To solve the above challenges, we are excited to announce the release of the novel LiteRT Tensor API , a lightweight authoring layer in the LiteRT stack, designed to improve handcrafting models.","rights":"original"}}]},{"id":"4c0bcf89-6329-463a-8853-00d9ad72c55d","kind":"capability","statement":"Developers can build computation graphs op by op in C++.","citations":[{"source":1,"supportProbability":0.99,"supports":true,"quote":null}]},{"id":"3c6cba36-25c0-4ebf-8ed2-8b4425c90a9e","kind":"capability","statement":"The API can load weights directly from checkpoints such as Hugging Face .safetensors files.","citations":[{"source":1,"supportProbability":0.99,"supports":true,"quote":{"text":"Direct weight ingestion : load weights straight from model checkpoints, such as Hugging Face .safetensors files, without conversion scripts needed in between.","rights":"original"}}]},{"id":"67fd72a2-2027-4883-8929-631f50cef057","kind":"deployment","statement":"The same graph can run through LiteRT on CPU, GPU, or NPU, or directly on CPU through XNNPACK.","citations":[{"source":1,"supportProbability":0.98,"supports":true,"quote":{"text":"The same graph can run through LiteRT on CPU, GPU (powered by open-source ML Drift ) and NPU, or without model file or runtime directly on CPU through XNNPACK, so you don't rewrite it for each new hardware target.","rights":"original"}}]},{"id":"a19535c3-f10e-469c-9e1a-722c484881a8","kind":"capability","statement":"Preprocessing, model partitions, and postprocessing can share accelerator buffers without copying intermediate data.","citations":[{"source":1,"supportProbability":0.97,"supports":true,"quote":{"text":"Tensor-centric buffer interop : pre-processing, model partitions, and post-processing pass tensors to each other through shared accelerator buffers, so intermediate data stays on the accelerator with zero copies.","rights":"original"}}]},{"id":"af0ea914-3395-4234-ac9b-63349fcbaf78","kind":"availability","statement":"Google provides a Tensor API guide, reference, and buildable end-to-end SAM 2 example.","citations":[{"source":1,"supportProbability":0.93,"supports":true,"quote":{"text":"Point it to the Tensor API README and the SAM 2 sample ; together they cover the API and a buildable end-to-end example.","rights":"original"}}]}]},"nextCursor":null,"license":{"catalog":{"name":"CC BY 4.0","spdx":"CC-BY-4.0","url":"https://creativecommons.org/licenses/by/4.0/","attribution":"wowsuchai","attributionUrl":"https://wowsuchai.withglint.app","covers":["catalog metadata","classifications","generated summaries","feed event metadata"]},"excluded":{"covers":["source quotations","linked source content"],"terms":"Not licensed by wowsuchai. Source quotations and linked source content keep their original rights."}}}