LetTokenBurnVideo-MME 10% Sample A 10% random sample of the Video-MME benchmark: 90 of the 900 videos, with their 270 multiple-choice questions and 78 subtitle files.
$ npm i -g lettokenburn-cli$ ltb bench info video-mme-sample
External harness. No Harbor adapter has been generated for this benchmark yet. `ltb bench info` prints the known links.
Runs happen on your machine with your own subscription; only the score summary is pushed. CLI docs
No sourced scores recorded yet.
Every number links to where it was published. Low-confidence rows are muted and need per-cell verification.
Video-MME 10% Sample
A 10% random sample of the Video-MME benchmark: 90 of the 900 videos, with their 270 multiple-choice questions and 78 subtitle files. It was prepared by Aintropy for quick, low-cost evaluation runs and pipeline smoke tests. This is not the original dataset and not an official release. The full benchmark, its paper and its leaderboard belong to the Video-MME authors:
Original dataset on Hugging Face: https://huggingface.co/datasets/lmms-eval/Video-MME Paper:…