Resources
Blog
Last updated — July 4, 2026
Product updates, engineering notes, and deep dives into LLM routing, performance, and developer experience.
July 2, 2026Engineering
Routing Claude Opus 4.8 alongside GLM and DeepSeek
How our gateway picks the right model for each request, and why Opus 4.8 changes the equation for agentic workflows.
6 min readJune 26, 2026Performance
Hitting 175 tps with GLM 5.2 Fast
Benchmarking throughput across providers, and the caching layer that keeps latency flat under load.
8 min readJune 18, 2026Product
Why one key beats seven
The case for a unified endpoint — billing, failover, and observability without the SDK sprawl.
4 min readJune 10, 2026Product
Instant keys via Cryptobot and xRocket
How Telegram-native payments let developers buy tokens without leaving the chat.
3 min readMay 28, 2026Engineering
Failover under the hood
What happens when a provider goes down, and how we keep your agents running.
7 min readMay 14, 2026Product
Per-key spending caps in practice
Preventing runaway costs when an agent loop goes off the rails.
5 min read