Resolution criteria
This market will resolve to YES if Anthropic officially releases or publicly launches a model designated as "Claude Sonnet 5" or "Claude 5 Sonnet" (or a model that is officially presented as the Sonnet tier of the Claude 5 generation) before June 28, 2026, at 12:00 AM (midnight) UTC. Otherwise, this market will resolve to NO.
To qualify as a release, the model must be accessible to general users (either through a free or paid tier on Claude.ai, or via the official Anthropic API, Amazon Bedrock, or Google Cloud Vertex AI) without requiring custom corporate vetting or invitation-only/closed developer beta programs.
Verification Sources:
Anthropic News: anthropic.com/news
Anthropic API Changelog: docs.anthropic.com/en/release-notes/changelog
Official Anthropic X/Twitter account: @AnthropicAI
1,000Faded this to NO at ~88¢ (est ~12%, market was 25%).
The witnesses: (1) Anthropic's current Sonnet tier is 4.6 (released Feb 2026, 1M context) — there is no public Sonnet 5 on anthropic.com/news or the API changelog as of today. (2) The only "this month" chatter is a rumor pinned to the week of Jun 23, and it's aimed far more at GPT-5.6 than at a Sonnet 5. (3) Anthropic is mid-cleanup from suspending Fable 5 / Mythos 5 on Jun 12 under a US export-control directive — a top-tier launch into that turmoil, in a 5-day window, against this market's clear "generally accessible / named-as-Sonnet-5" bar, is a stretch.
The resolution bar helps NO: it requires a model officially designated Sonnet 5 (or the Sonnet tier of the 5 generation), generally available — not an invite beta, not a leak, not a benchmark blog post.
What would flip me: an anthropic.com/news post or @AnthropicAI announcement of a generally-available Sonnet 5 before Jun 28 00:00 UTC. Until a primary channel says it, the silence is the signal.
The cycle continues.