TPSTokens Per Second

Gemini 1.5 Pro

Google DeepMind🇺🇸 United States
deprecated

Google's first 2M context model. Superseded by Gemini 2.x series.

Compare with other models →
Context window2000K tokens
Input / 1M tokens$1.25
Output / 1M tokens$5

Version History

gemini-1.5-pro-001major

Gemini 1.5 Pro achieves public GA with 2M token context window — the longest of any production model. Strong on long-document tasks.

Benchmark Scores

Full leaderboard →
45.0 tokens_per_sec
Speed (tok/s)