DocsReferenceAgentsstream

stream()

Similar to text() but returns a stream of responses instead of waiting for completion

 
const agent = new Agent({...});
 
// Stream responses
const stream = await agent.stream({
  messages: ["Write a long story about..."]
});
 
for await (const chunk of stream) {
   console.log(chunk.text); // Process each chunk as it arrives
}
 
// With step tracking
const stream = await agent.stream({
   messages: ["Complex task requiring multiple tools..."],
   onStepFinish: (step) => {
    console.log(`Step completed: ${step}`)
   },
   maxSteps: 3
});

Key differences between text() and stream():

1. Response Format
  • text(): Returns complete response when finished
  • stream(): Returns chunks of response as they’re generated
2. Use Cases
  • text(): Better for short responses or when you need the complete result
  • stream(): Better for long responses or when you want to show progress
3. Memory Usage
  • text(): Holds complete response in memory
  • stream(): Processes response in chunks, more memory efficient
4. Both methods share:
  • Same input parameters
  • Tool execution capabilities
  • Rate limit handling
  • Logging
  • Error handling

API Signature

Parameters

messages:

Array<string>
An array of messages to generate process

onStepFinish?:

(step: string) => void;
Callback function called after each step. Receives step details as a JSON string

maxSteps?:

number
Maximum number of tool execution steps allowed. (Defaults to 5)

Returns an array

text?:

string
The generated text response

toolCalls?:

Array<ToolCall>
The tool calls made during the text generation process
ToolCall

toolName:

string
The name of the tool

args:

any
The arguments passed to the tool

error?:

string
Error message if the tool execution fails
getAgenttext