Post by Trek10, Inc.

Anthropic Claude Sonnet 4.6 recently launched, and it delivers enterprise-grade coding at Sonnet pricing! With a jump to 72.5% on OSWorld-Verified, meaningful gains on SWE-bench and Terminal-Bench, and no increase in base pricing, it’s now the best general-purpose model in terms of price-performance. In a new blog by Guille Ojeda, we break down what the benchmarks actually mean, where Sonnet 4.6 closes the gap, how prompt injection resilience improved, and how to think about cost, thinking tokens, and long-context pricing before you deploy. Learn more 👉 https://hubs.li/Q04544kQ0 Trek10 is now part of Caylent! Follow Caylent for future updates and news.