Capabilities & practical fit
Coding & agents
Positioned as a subagent for narrow coding tasks; Anthropic recommends larger models for complex agentic coding.
Routing & classification
Targets high-volume classification, summaries and quick database queries.
Output & integration
Generative responses, with browser and computer use support described in beta SDK updates.
Editorial fit
Editorial interpretation: evaluate for repetitive, latency-sensitive work rather than assuming flagship-level performance on every task.
Capability descriptions reflect provider positioning. Editorial fit is an interpretation, not a measured benchmark.
Compare this model with another →