5 ms·
We've done extensive comparisons against GPT-4V for video inputs in our technical report: https://storage.googleapis.com/deepmind-media/gemini/gemini_v1_5_repor
by ankeshanand 3y ago
We've done extensive comparisons against GPT-4V for video inputs in our technical report: https://storage.googleapis.com/deepmind-media/gemini/gemini_v1_5_report.pdf https://storage.googleapis.com/deepmind-media/gemini/gemini_....
Most notably, at 1FPS the GPT-4V API errors out around 3-4 mins, while 1.5 Pro supports upto an hour of video inputs.
- jxy 3y agoSo that 3-4 mins at 1FPS means you are using about 500 to 700 tokens per image, which means you are using `detail: high` with something like 1080p to feed to gpt-4-vision-preview (unless you have another private endpoint). The gemini 1.5 pro uses about 258 tokens per frame (2.8M tokens for 10856 frames). Are those comparable?
- verticalscaler 3y agoThe average shot length in modern movies is between 4 and 16 seconds and around 1 minute for a scene.
- moralestapia 3y ago>while 1.5 Pro supports upto an hour of video inputs At what price, tho?