Compare models

Context, limits, capabilities, price and use cases side by side — up to 4 models at a time. The URL holds your selection, so you can share it.

Gemini 2.5 FlashGemini 2.5 ProGemini 2.5 Flash Lite Add a modelChat with these 3

AttributeHide identical rows	Gemini 2.5 Flashgemini/gemini-2.5-flash	Gemini 2.5 Progemini/gemini-2.5-pro	Gemini 2.5 Flash Litegemini/gemini-2.5-flash-lite
Specs
Context window	1M	1M	1M
Max output	66K	66K	66K
Input modalities	TextImageAudioVideo	TextImageAudioVideo	TextImageAudioVideo
Output modalities	Text	Text	Text
Capabilities
Streaming	Supported	Supported	Supported
Tool calling	Supported	Supported	Supported
Structured output (JSON)	Supported	Supported	Supported
Image input	Supported	Supported	Supported
Thinking mode	Supported	Supported	Supported
Price / 1M tokens
Input	$ 0.30	$ 1.25>= 200K$ 2.50	$ 0.10
Output	$ 2.50	$ 10.00>= 200K$ 15.00	$ 0.40
Cache read	$ 0.030	$ 0.13>= 200K$ 0.25	$ 0.010
Cache write	$ 0.30	$ 1.25>= 200K$ 2.50	$ 0.10
What it is good at
Best for	High-volume reasoning Multimodal analysis	Long-context analysis Math and STEM reasoning	Classification and extraction Fast translation
Published benchmarks
SWE-Bench Verifiedresolved_ratesetups differ by: harness, reasoning	60.4%enabled(setup not fully disclosed)	63.8%custom-agent(setup not fully disclosed)	27.6%enabled(setup not fully disclosed)
AIME 2025accuracy	72%enabled(setup not fully disclosed)	88%enabled(setup not fully disclosed)	63.1%enabled(setup not fully disclosed)
MMMUaccuracysetups differ by: reasoning	79.7%enabled(setup not fully disclosed)	84%(setup not fully disclosed)	72.9%enabled(setup not fully disclosed)
Humanity's Last Examaccuracysetups differ by: reasoning	11%enabled(setup not fully disclosed)	18.8%(setup not fully disclosed)	6.9%enabled(setup not fully disclosed)
GPQA Diamondaccuracy	82.8%enabled(setup not fully disclosed)	86.4%enabled(setup not fully disclosed)	66.7%enabled(setup not fully disclosed)