Anthropic |
AI provider implementation for the Anthropic Messages REST API (Claude models),
with automatic retry on rate limit (429) and overload (529) responses.
Inheritance HierarchyNamespace: Seal.AI
Assembly: SealLibrary (in SealLibrary.dll) Version: 10.0.1.0+a44cd3a7f03c2bf3cdbd971a878c679e40bed08a
SyntaxC#
public class AnthropicProvider : AIProvider
The AnthropicProvider type exposes the following members.
Constructors| Name | Description | |
|---|---|---|
 | AnthropicProvider | Initializes a new Anthropic provider. |
Properties| Name | Description | |
|---|---|---|
 | GUID | Unique identifier for this provider instance. (Inherited from AIProvider) |
| LastUsage |
Token usage of the last API call (input/output tokens, one call).
Zeroes when the API did not report usage information.
(Inherited from AIProvider) |
Methods| Name | Description | |
|---|---|---|
| Equals | Determines whether the specified object is equal to the current object. (Inherited from Object) |
 | Finalize | Allows an object to try to free resources and perform other cleanup operations before it is reclaimed by garbage collection. (Inherited from Object) |
| GetHashCode | Serves as the default hash function. (Inherited from Object) |
| GetType | Gets the Type of the current instance. (Inherited from Object) |
| HandleChat |
Sends messages to the AI and returns the agent reply.
The reply is automatically appended to messages.
(Overrides AIProviderHandleChat(ListChatMessage)) |
| HandleChatWithTools |
Sends messages together with available tools to the AI.
If the model decides to invoke one or more tools the descriptors are placed in
toolCalls and the method returns Empty;
the agent message (with embedded tool-call data) is appended to messages
automatically so the conversation can be continued after the caller executes the tools.
If the model produces a text reply the reply is appended to messages and returned.
(Overrides AIProviderHandleChatWithTools(ListChatMessage, IListAITool, IListAIToolCall)) |
| Initialize |
Configures the provider with the endpoint URL, API key, model name, and generation parameters.
Called internally by GetAIProvider(ProviderType, String, String, String, Single, Int32, Single) immediately after construction.
(Overrides AIProviderInitialize(String, String, String, Single, Int32, Single)) |
| MemberwiseClone | Creates a shallow copy of the current Object. (Inherited from Object) |
| SetLastUsage |
Records the token usage of the API call that just completed.
(Inherited from AIProvider) |
| ToString | Returns a string that represents the current object. (Inherited from Object) |
Fields| Name | Description | |
|---|---|---|
 | _apiKey | API key used to authenticate against the provider. (Inherited from AIProvider) |
| _endpoint | Resource endpoint URL of the provider API. (Inherited from AIProvider) |
| _maxTokens | Maximum number of tokens to generate (0 = provider default). (Inherited from AIProvider) |
| _model | Name of the model to use (e.g. gpt-4o, claude-3-5-sonnet-20241022, llama3). (Inherited from AIProvider) |
| _temperature | Sampling temperature (0.0 = default, deterministic). (Inherited from AIProvider) |
| _topP | Nucleus sampling probability (1.0 = default, disabled). (Inherited from AIProvider) |
See Also