Anthropic Makes Claude Sonnet 5 the Default, Pushing Near-Frontier Agents to a Fraction of Opus Pricing
Sonnet 5 now ships as the default model across free and Pro plans, arriving with a 1M-token context window and agentic performance Anthropic positions close to Opus 4.8 — at promotional pricing of $2 per million input tokens through August 31.
Anthropic released Claude Sonnet 5 on July 2 and immediately made it the default model across every plan tier, a move that reframes what "mid-tier" means in the current model hierarchy. According to a detailed briefing from @whytrend_ai, Sonnet 5 delivers autonomous-execution performance approaching the higher-end Opus 4.8, but at introductory pricing of $2 per million input tokens and $10 per million output tokens through August 31. That is roughly half the cost readers would associate with near-frontier capability, and the pricing is explicitly promotional — a detail worth planning around before the window closes.
The more consequential change for developers may be quieter. As @whytrend_ai reported, Claude Code shipped in version v2.1.197 with Sonnet 5 as its default model and a 1M-token long-context window now bundled in as standard. For anyone running coding agents against large repositories, that is a structural shift: a million-token window standard-issue means fewer retrieval hacks, fewer context-window workarounds, and longer autonomous sessions before an agent loses the thread.
Get our free daily newsletter
Get this article free — plus the lead story every day — delivered to your inbox.
Want every article and the full archive? Upgrade anytime.
No spam. Unsubscribe anytime.