Class: Anthropic
Anthropic LLM implementation
Implements
Constructors
constructor
• new Anthropic(init?)
Parameters
| Name | Type |
|---|---|
init? | Partial<Anthropic> |
Defined in
packages/core/src/llm/LLM.ts:667
Properties
apiKey
• Optional apiKey: string = undefined
Defined in
packages/core/src/llm/LLM.ts:660
callbackManager
• Optional callbackManager: CallbackManager
Defined in
packages/core/src/llm/LLM.ts:665
hasStreaming
• hasStreaming: boolean = true
Implementation of
Defined in
packages/core/src/llm/LLM.ts:651
maxRetries
• maxRetries: number
Defined in
packages/core/src/llm/LLM.ts:661
maxTokens
• Optional maxTokens: number
Defined in
packages/core/src/llm/LLM.ts:657
model
• model: "claude-2" | "claude-instant-1"
Defined in
packages/core/src/llm/LLM.ts:654
session
• session: AnthropicSession
Defined in
packages/core/src/llm/LLM.ts:663
temperature
• temperature: number
Defined in
packages/core/src/llm/LLM.ts:655
timeout
• Optional timeout: number
Defined in
packages/core/src/llm/LLM.ts:662
topP
• topP: number
Defined in
packages/core/src/llm/LLM.ts:656
Accessors
metadata
• get metadata(): Object
Returns
Object
| Name | Type |
|---|---|
contextWindow | number |
maxTokens | undefined | number |
model | "claude-2" | "claude-instant-1" |
temperature | number |
tokenizer | undefined |
topP | number |
Implementation of
Defined in
packages/core/src/llm/LLM.ts:691
Methods
chat
▸ chat<T, R>(messages, parentEvent?, streaming?): Promise<R>
Get a chat response from the LLM
Type parameters
| Name | Type |
|---|---|
T | extends undefined | boolean = undefined |
R | T extends true ? AsyncGenerator<string, void, unknown> : ChatResponse |
Parameters
| Name | Type | Description |
|---|---|---|
messages | ChatMessage[] | The return type of chat() and complete() are set by the "streaming" parameter being set to True. |
parentEvent? | Event | - |
streaming? | T | - |
Returns
Promise<R>
Implementation of
Defined in
packages/core/src/llm/LLM.ts:719
complete
▸ complete<T, R>(prompt, parentEvent?, streaming?): Promise<R>
Get a prompt completion from the LLM
Type parameters
| Name | Type |
|---|---|
T | extends undefined | boolean = undefined |
R | T extends true ? AsyncGenerator<string, void, unknown> : ChatResponse |
Parameters
| Name | Type | Description |
|---|---|---|
prompt | string | the prompt to complete |
parentEvent? | Event | - |
streaming? | T | - |
Returns
Promise<R>
Implementation of
Defined in
packages/core/src/llm/LLM.ts:776
mapMessagesToPrompt
▸ mapMessagesToPrompt(messages): string
Parameters
| Name | Type |
|---|---|
messages | ChatMessage[] |
Returns
string
Defined in
packages/core/src/llm/LLM.ts:702
streamChat
▸ Protected streamChat(messages, parentEvent?): AsyncGenerator<string, void, unknown>
Parameters
| Name | Type |
|---|---|
messages | ChatMessage[] |
parentEvent? | Event |
Returns
AsyncGenerator<string, void, unknown>
Defined in
packages/core/src/llm/LLM.ts:751
streamComplete
▸ Protected streamComplete(prompt, parentEvent?): AsyncGenerator<string, void, unknown>
Parameters
| Name | Type |
|---|---|
prompt | string |
parentEvent? | Event |
Returns
AsyncGenerator<string, void, unknown>
Defined in
packages/core/src/llm/LLM.ts:794
tokens
▸ tokens(messages): number
Calculates the number of tokens needed for the given chat messages
Parameters
| Name | Type |
|---|---|
messages | ChatMessage[] |
Returns
number