{"name":"Looking-Ahead","repo":"TencentARC/Looking-Ahead","url":"https://midorreal.com/project/tencentarc-looking-ahead","repository":"https://github.com/TencentARC/Looking-Ahead","description":"Looking Ahead: Modeling Future Evolution in Autoregressive Video Generation","facts":{"language":{"value":"Python","source":"https://github.com/TencentARC/Looking-Ahead","checked":"21 Nov 2026"},"homepage":{"value":"https://hellorpg.github.io/looking-ahead/","source":"https://hellorpg.github.io/looking-ahead/","checked":"21 Nov 2026"},"releases":{"value":"0","source":"https://github.com/TencentARC/Looking-Ahead/releases","checked":"20 Nov 2026"},"contributors":{"value":"2","source":"https://github.com/TencentARC/Looking-Ahead/graphs/contributors","checked":"20 Nov 2026"},"archived":{"value":"no","source":"https://github.com/TencentARC/Looking-Ahead","checked":"21 Nov 2026"}},"stars":275.0,"writeup":{"what_it_is":"Looking Ahead is a research project on autoregressive video generation. It introduces Future Chunk Prediction (FCP), which trains video models to predict temporally distant video chunks from intermediate features of the current chunk.","audience":"Researchers and developers working on autoregressive video generation and world models","claims":[{"kind":"broad","claim":"FCP learns more semantically structured representations","excerpt":"learns more semantically structured representations","status":"too_broad"},{"kind":"specific","claim":"Released weights for 1.3B and 5B text-and-image-to-video models","excerpt":"weights for 1.3B and 5B autoregressive text-and-image-to-video (TI2V) models","status":"not_checked"},{"kind":"specific","claim":"Tested on Linux with Python 3.12 and PyTorch 2.10.0 with CUDA 12.8","excerpt":"Tested on Linux with Python 3.12 and PyTorch 2.10.0+cu128 (CUDA 12.8).","status":"not_checked"},{"kind":"specific","claim":"Generates a 5-second video from an input image and prompt on a single GPU","excerpt":"It generates a 5-second video using the 1.3B FCP model","status":"not_checked"}],"alternatives":[],"written_by":"AI, from the project's README","date":"2026-10-28"},"why_now":null,"owner_supplied":null,"score":{"verdict":"sleeper","hype":19,"reality":50,"gap":-31,"momentum":null,"confidence":60,"version":"score-2.0","rule":"verdict-2.0","calculated_at":"2026-11-22T08:00:00+00:00","verdict_source":"rule"},"early_read":null,"changes":[{"date":"2026-10-21T11:20:59+00:00","text":"Repository created","source":"https://github.com/TencentARC/Looking-Ahead"}],"cite":"Looking-Ahead is a GitHub project at https://github.com/TencentARC/Looking-Ahead written mainly in Python, described there as \"Looking Ahead: Modeling Future Evolution in Autoregressive Video Generation\".","attribution":"Data from Mid or Real, https://midorreal.com/project/tencentarc-looking-ahead, checked 22 November 2026. Attribution with a link is required.","still_real":null}