Runner Configuration

Enable auto-save for user input blobs

When enabled, the runner automatically saves any incoming inlineData parts from the user's message as artifacts before appending the event. Each saved artifact is named artifact_<invocationId>_<index> and the original message part is replaced with a short text placeholder.

Requirements:

An artifactService must be configured on the Runner
Enable via withRunConfig({ saveInputBlobsAsArtifacts: true }) on the agent builder, or per-run via RunConfig

import {
  agentBuilder,
  Runner,
  InMemorySessionService,
  InMemoryArtifactService,
  RunConfig
} from '@iqai/adk';

// Configure at agent build time
const agent = agentBuilder()
  .name('artifact_agent')
  .model('gemini-2.5-flash')
  .withRunConfig({ saveInputBlobsAsArtifacts: true })
  .build();

const runner = new Runner({
  appName: 'my_app',
  agent,
  sessionService: new InMemorySessionService(),
  artifactService: new InMemoryArtifactService(),
});

// Or enable per run
// await runner.runAsync({
//   userId,
//   sessionId,
//   newMessage,
//   runConfig: new RunConfig({ saveInputBlobsAsArtifacts: true })
// });

Artifacts saved by this mechanism use the incoming part's inlineData (base64 string + MIME type) as-is. Ensure your front-end encodes uploads as base64 and sets a correct mimeType.

Minimal example: sending an image upload

import type { Content } from '@google/genai';

const imageBytes = Buffer.from(/* raw PNG bytes */);
const message: Content = {
  role: 'user',
  parts: [
    {
      inlineData: {
        data: imageBytes.toString('base64'),
        mimeType: 'image/png'
      }
    }
  ]
};

for await (const event of runner.runAsync({ userId: 'u1', sessionId: 's1', newMessage: message })) {
  // Events stream
}

// The image will be saved as an artifact, e.g. "artifact_<invocationId>_0"

Runner Configuration

Enable auto-save for user input blobs

Minimal example: sending an image upload

On this page