Google: Gemini 2.5 Flash Preview 05-20

Text input Image input File input Text output Unavailable
Author's Description

Gemini 2.5 Flash May 20th Checkpoint is Google's state-of-the-art workhorse model, specifically designed for advanced reasoning, coding, mathematics, and scientific tasks. It includes built-in "thinking" capabilities, enabling it to provide responses with greater accuracy and nuanced context handling. Note: This model is available in two variants: thinking and non-thinking. The output pricing varies significantly depending on whether the thinking capability is active. If you select the standard variant (without the ":thinking" suffix), the model will explicitly avoid generating thinking tokens. To utilize the thinking capability and receive thinking tokens, you must choose the ":thinking" variant, which will then incur the higher thinking-output pricing. Additionally, Gemini 2.5 Flash is configurable through the "max tokens for reasoning" parameter, as described in the documentation (https://openrouter.ai/docs/use-cases/reasoning-tokens#max-tokens-for-reasoning).

Key Specifications
Cost
$$$
Context
1M
Released
May 20, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Tools Structured Outputs Tool Choice Reasoning Include Reasoning Response Format Stop Top P Max Tokens Temperature
Features

This model supports the following features:

Tools Reasoning Response Format Structured Outputs
Performance Summary

Google's Gemini 2.5 Flash Preview 05-20 demonstrates strong performance as a state-of-the-art workhorse model, particularly excelling in advanced reasoning, coding, mathematics, and scientific tasks. It consistently performs among the fastest models, ranking in the 82nd percentile across benchmarks, and offers competitive pricing, typically falling within the 69th percentile. A standout feature is its exceptional reliability, achieving a 100% success rate across all benchmarks, indicating consistent and usable responses. In specific benchmark categories, Gemini 2.5 Flash achieved perfect accuracy in Ethics, notably being the most accurate model at its price point and among models of comparable speed. It also showed high accuracy in Email Classification (99.0%), proving to be the most accurate among models of similar speed. While its General Knowledge accuracy was strong at 99.0%, its Coding performance, at 83.0% accuracy, was more moderate compared to other categories. The model's configurable "thinking" capability, with distinct pricing for its "thinking" variant, allows for nuanced context handling and greater accuracy, though this comes at a higher cost. Its primary strength lies in its reliability and ethical reasoning, coupled with competitive speed and cost-effectiveness for many tasks.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.15
Completion $0.6
Input Cache Read $0.0375
Input Cache Write $0.233

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Google
Google | google/gemini-2.5-flash-preview-05-20 1M $0.15 / 1M tokens $0.6 / 1M tokens
Google AI Studio
Google AI Studio | google/gemini-2.5-flash-preview-05-20 1M $0.15 / 1M tokens $0.6 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by google