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 ProGemini 2.5 Flash Add a modelChat with these 2

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