Gemini 1.5 Pro was released February 15, 2024 with a breakthrough 1M token context window (later extended to 2M). Used a Mixture-of-Experts architecture for efficiency. Could process up to 1 hour of video, 11 hours of audio, or 30,000+ lines of code. Priced at \$1.25/\$5 per million tokens.
Modality
text, image, audio, video
Capabilities4
tool-usevisionaudio-inputlong-context