Model optimization
Gemini Flash compression
Gemini 1.5 Flash offers a 1M token context window at competitive pricing. The large context window invites bloat — compression keeps costs down.
Flash context costs
Gemini 1.5 Flash costs $0.075/1M input tokens for contexts up to 128K tokens, and $0.15/1M for longer contexts. With a 100K-token context, that is $0.0075 per query. Compression to 35K tokens drops this to $0.0026 per query.
Frequently asked questions
Does the large context window make compression less important?
The opposite — a 1M window invites more context, making compression more valuable.
Does compression work with Gemini's multimodal inputs?
Compression applies to the text portions of multimodal inputs.
Build with less context
Put compression in front of your next LLM call.
Use the hosted API or run SuperCompress locally. Keep the evidence, drop the token waste, and measure the savings before it reaches your model.