Quick answer
Anthropic released Claude Sonnet 5.5 on September 28, 2026. The API model ID is claude-sonnet-5-5; Anthropic lists input and output prices of $2 and $10 per million tokens, the same as Sonnet 5, while reporting up to 30% lower cost per task and 30%+ faster generation in its own tests. The model is available through Anthropic's listed platform and cloud routes. Results are provider-reported, migration details matter, and availability in Chat AI is not verified.
What Claude Sonnet 5.5 is positioned to do
Anthropic released Sonnet 5.5 on September 28 as the second model in its 5.5 family. The company positions it for well-scoped everyday work such as fixing bugs and creating documents, slides, and spreadsheets, while Opus 5.5 is aimed at complex work requiring sustained judgment. Anthropic reports gains over Sonnet 5 in its evaluations, but these are provider-reported results rather than an independent Chat AI comparison.
Sources: Anthropic
Per-token pricing is unchanged; Anthropic reports lower cost per task
Anthropic lists Sonnet 5.5 at $2 per million input tokens and $10 per million output tokens, the same rates it lists for Sonnet 5. It says Sonnet 5.5 generates output 30% or more faster and costs up to 30% less per task for most work because it typically uses fewer tokens. Those speed and task-cost comparisons come from Anthropic's own tests; actual usage and bills depend on the task, output length, and configuration.
Sources: Anthropic
Access includes the Claude Platform and named cloud providers
Anthropic says Sonnet 5.5 is available on its platforms, including Amazon Web Services, Google Cloud, and Microsoft Azure. The Claude Platform API model ID is claude-sonnet-5-5. Existing API integrations should review the launch migration note: applications that run Sonnet with thinking disabled need to move to the between_tools setting to keep up-front thinking off.
Sources: Anthropic
Some higher-risk requests can be blocked or switched to Sonnet 5
Anthropic says Sonnet 5.5 has stronger cybersecurity capabilities than Sonnet 5 and uses classifiers for a narrow set of higher-risk requests. Depending on the classifier, a request may be blocked or, when automatic switching is enabled, rerun on Sonnet 5; the Help Center says these switches are labeled in Claude. Routine coding is not the target of these safeguards, but legitimate specialized work may still encounter a false positive. Anthropic's biology safeguards remain the same as Sonnet 5's.
Sources: Anthropic, Anthropic Help Center
Sonnet 5.5 availability in Chat AI is not verified
Chat AI's published model directory does not show an exact Claude Sonnet 5.5 entry. Do not assume the model is included in the app; check the current model selector before relying on it. The directory's model records were last verified on August 24, before Anthropic announced this release.
Sources: Chat AI
Frequently asked questions
What readers usually ask
What is Claude Sonnet 5.5's API model ID?
Anthropic lists claude-sonnet-5-5 for the Claude Platform API. Check the provider documentation for the current identifier and integration details before deploying.
Did Anthropic lower Sonnet's per-token API price?
No. Anthropic lists $2 per million input tokens and $10 per million output tokens, the same as Sonnet 5. It reports up to 30% lower cost per task for most work because Sonnet 5.5 typically uses fewer tokens.
Is Claude Sonnet 5.5 available in Chat AI?
Its availability is not verified. The published Chat AI model directory does not contain an exact Sonnet 5.5 entry; check the live app's model selector rather than assuming access.
Does Sonnet 5.5 always answer higher-risk cybersecurity requests?
No. Anthropic says a narrow set of higher-risk requests may be blocked or switched to Sonnet 5 when automatic switching is active. The Help Center describes the fallback behavior and says switches are labeled.
Evidence