SilentAPI

Increased Rate Limits

To better support scaling applications, we have increased the rate limits for several of our most popular models.

• » GPT-5: 25 → 30/minute
• » GPT-5 Mini: 50 → 60/minute
• » GPT-5 Nano: 100 → 120/minute
• » Gemini 2.5 Flash: 10 → 30/minute
• » Gemini 2.5 Flash Lite: 50 → 60/minute
• » Gemini 2.0 Flash: 25 → 30/minute
• » Gemini 2.0 Flash Lite: 50 → 60/minute

Price Reductions

We are implementing price reductions on the following key services:

• Pricing for all Gemini 2.5 models has been reduced by up to 25% for Output and Thinking tokens.

We anticipate further significant price decreases in version 0.3 as we continue to optimize our infrastructure and operational costs.

New Features & Improvements

GET Requests for GPT Models: You can now send simple text-only queries to GPT models using GET requests, making integration easier and faster.

Pricing Fixes: We’ve addressed a few issues to make cost calculations more accurate.

• Resolved an issue where “thinking” token costs for Gemini 2.5 models were miscalculated. You can now manage these costs by disabling thinking or setting a thinking budget.
• Fixed a billing issue affecting users with low balances to ensure accurate AI usage charges.

Mobile Layout: The Services, Billing, and Metrics pages now adapt better to mobile screens for a smoother experience.

Other Updates: This release includes small bug fixes, visual tweaks, and performance improvements across the platform.

Introducing Google Vision Tools

We are pleased to introduce a suite of powerful Google Vision tools. These are now available under Services → Media:

• » Document Text Detection
• » Face Detection
• » Image Properties Detection
• » Label Detection
• » Landmark Detection
• » Logo Detection
• » Safe Search Detection
• » Text Detection
• » Web Detection
• » Object Localization