Gemini 2.5 Flash
Google · Released May 20, 2025
$0.3
Input price / MTok
#8 of 40$2.5
Output price / MTok
#10 of 401.0M
Context window
#23 of 41222
Tokens / sec
#1 of 2348.9
SWE-bench Verified
#12 of 1282.8
GPQA Diamond
#13 of 1972
AIME 2025
#10 of 1059.3
LiveCodeBench
#7 of 8Summary
Gemini 2.5 Flash is a proprietary language model from Google, released on May 20, 2025.
At $0.3 input and $2.5 output per million tokens, it ranks #8 of 40 for input price, placing it mid-range on input cost.
Its context window of 1.0M tokens ranks #23 of 41 for context depth in this dataset.
It targets budget, coding, agents workloads.
Throughput is measured at 222 tokens per second (ranked #1 of 23 for throughput).
First Flash model with thinking capabilities.
Benchmarks
| Benchmark | Score | Max | % | Rank | Source |
|---|---|---|---|---|---|
| SWE-bench Verified | 48.9 | 100 | 48.9% | #12/12 | Google DeepMind model card (self-reporte... |
| GPQA Diamond | 82.8 | 100 | 82.8% | #13/19 | Google DeepMind model card (self-reporte... |
| AIME 2025 | 72 | 100 | 72.0% | #10/10 | Google DeepMind model card (self-reporte... |
| LiveCodeBench | 59.3 | 100 | 59.3% | #7/8 | Google DeepMind model card (self-reporte... |
Similar Models
Compare all 6 modelsModels with the most shared benchmark coverage and closest scores to Gemini 2.5 Flash.
Compare Gemini 2.5 Flash side by side
Pre-filled with 5 similar models
Speed source: https://artificialanalysis.ai/models/gemini-2-5-flash