模型
Models
class CapabilitySupport { $supported; }
Whether this capability is supported by the model.
class ContextManagementCapability { $clearThinking20251015; $clearToolUses20250919; $compact20260112; $supported; }
Indicates whether a capability is supported.
Indicates whether a capability is supported.
Indicates whether a capability is supported.
Whether this capability is supported by the model.
class EffortCapability { $high; $low; $max; /* 3 more */ }
Whether the model supports high effort level.
Whether the model supports low effort level.
Whether the model supports max effort level.
Whether the model supports medium effort level.
Whether this capability is supported by the model.
Indicates whether a capability is supported.
class ModelCapabilities { $batch; $citations; $codeExecution; /* 6 more */ }
Whether the model supports the Batch API.
Whether the model supports citation generation.
Whether the model supports code execution tools.
Context management support and available strategies.
Effort (reasoning_effort) support and available levels.
Whether the model accepts image content blocks.
Whether the model accepts PDF content blocks.
Whether the model supports structured output / JSON mode / strict tool schemas.
Thinking capability and supported type configurations.
class ModelInfo { $type = 'model'; $id; $capabilities; /* 4 more */ }
"model" typeObject type.
Object type.
For Models, this is always "model".
Unique model identifier.
Model capability information.
RFC 3339 datetime string representing the time at which the model was released. May be set to an epoch value if the release date is unknown.
A human-readable name for the model.
Maximum input context window size in tokens for this model.
Maximum value for the max_tokens parameter when using this model.
class ThinkingCapability { $supported; $types; }
Whether this capability is supported by the model.
Supported thinking type configurations.
class ThinkingTypes { $adaptive; $enabled; }
Whether the model supports thinking with type 'adaptive' (auto).
Whether the model supports thinking with type 'enabled'.