Models API Now Returns Capability Metadata

Anthropic has enhanced the Models API with structured capability information. Both GET /v1/models and GET /v1/models/{model_id} endpoints now return three new fields:

• max_input_tokens: Maximum input context length for the model
• max_tokens: Maximum output token limit
• capabilities: Object describing supported features (extended thinking, vision, tool use, etc.)

Developer Benefits

Rather than consulting documentation or maintaining hardcoded assumptions about model capabilities, developers can now query the API at runtime to discover:

• Which models support extended thinking or adaptive thinking
• Context window sizes for each model variant
• Media processing capabilities (vision, PDF handling)
• Tool use and function calling support
• Any other feature capabilities

This enables more dynamic, future-proof integrations that automatically adapt as new models and features become available.