Vertex
Claude Sonnet 4.6 brings improved coding performance, expanded developer tools, enhanced API capabilities, and updated safety evaluations.
Model details
Claude Sonnet 4.6 marks the next step in Anthropic's Sonnet family, arriving as a refined successor focused on practical developer workloads. Coverage from Help Net Security describes the release as bringing improved coding performance, expanded developer tools, enhanced API capabilities, and updated safety evaluations, signaling that the model targets real engineering tasks rather than headline benchmarks alone. Anthropic promoted this version as its most capable Sonnet to date, making it the default across claude.ai and Cowork while keeping pricing unchanged for users stepping up from the prior generation.
Independent reporting from YourStory highlights wider skills, long-context handling, and developer-facing upgrades as the defining improvements in this release, framing Sonnet 4.6 as a workhorse for teams building assistants, agents, and code-aware tools. Early third-party technical writeups appeared within days of launch, reflecting strong developer interest in testing its coding and API behavior. For teams choosing a hosted Claude option through Vertex, the combination of better coding results, broader developer tooling, and long-context support positions this model as a balanced choice for production applications that need both reasoning depth and reliable integration features.
@ai-sdk/google-vertex/anthropicTransparent token rates
Rates are shown per one million tokens. Combined means one million input plus one million output tokens.
Vertex
Claude Sonnet 4.6 brings improved coding performance, expanded developer tools, enhanced API capabilities, and updated safety evaluations.
Vertex
The company makes its most capable Sonnet model the default across claude.ai and Cowork, keeps pricing unchanged, and rolls out new tooling in the API.
This exact model name is also listed by 16 other providers.