Skip to content

Evaluation

Video needle-in-a-haystack retrieval (long-context)

Cross-modal needle-in-a-haystack: retrieve 'the secret word' text overlaid on a random frame of a 10.5-hour video (7 copies of AlphaGo documentary, 37,994 frames, 9.9M tokens)

Multimodal Google DeepMind — Gemini

How it was run

Open a release to see the setup, scoring, and other details that the lab published. Only fields the lab actually disclosed are shown.

Gemini 1.5
Google DeepMind — Gemini

rich detail 10 method fields published

See how this test was run