What's New

Gain deep insights into your AI usage with our new analytics dashboard.

New features:

• Usage Analytics — Detailed charts showing token consumption, request counts, and cost breakdown by model over time.
• Cost Breakdown — See exactly how much you're spending on each model, with comparison to what you'd pay at official pricing.
• Export to CSV — Download your usage data for internal reporting and reconciliation.
• Budget Alerts — Set custom thresholds and receive email notifications when your spend approaches your limit.

Other improvements:

• API key management now supports custom names and tags for easier organization.
• Added auto-recharge: set a threshold and your account tops up automatically when your balance runs low.
• Performance improvements: median API latency reduced by 12% through infrastructure optimizations.

This release significantly expands our model catalog.

New models added:

• Gemini 2.5 Pro — Google's most capable model with a 1 million token context window.
• Gemini 2.5 Flash — Lightning-fast responses at incredibly competitive pricing.
• DeepSeek V3 — Open-source powerhouse for code generation and reasoning tasks.
• DeepSeek R1 — Specialized reasoning model with chain-of-thought capabilities.

Improvements:

• Model status page now shows real-time latency and availability for each provider.
• Added model capability tags (vision, function calling, JSON mode) to the pricing page.
• Improved error messages when a model is temporarily unavailable, with automatic fallback suggestions.

We're thrilled to announce the official launch of TokenFast!

Core features in this release:

• Unified API Gateway — Access Claude, GPT, and Gemini models through a single OpenAI-compatible API endpoint.
• Instant API Key Provisioning — Purchase a plan and receive your API key in seconds, fully automated.
• Real-time Usage Dashboard — Monitor your token consumption, costs, and request volume in real time.
• Stripe-powered Billing — Secure payment processing with automatic receipt generation.
• 10% Below Official Pricing — Every model priced at a 10% discount compared to the provider's official rate.

We've been running a closed beta for the past three months and are confident in the platform's stability. Thank you to all our early testers for the invaluable feedback!