Interactions API: Breaking changes migration guide (May 2026)

The v1beta Interactions API is introducing breaking changes that restructure the API shape to support future capabilities like mid-flight steering and asynchronous tool calls. This page explains what's changing and provides before-and-after code examples to help you migrate. There are two categories of changes:

  1. Steps schema: A new steps array replaces the outputs array, providing a structured timeline of each interaction turn.
  2. Output format configuration: A new polymorphic response_format consolidates all output format controls and removes response_mime_type.

Follow the steps in How to migrate to the new schema to update your integration.

Core change: outputs to steps

The new schema replaces the outputs array with a steps array.

  • Legacy: Responses returned a flat outputs array containing only the model's generated content.
  • New schema: Responses return a steps array containing structured steps with type discriminators and per-step status fields.

POST /interactions returns only output steps. GET /interactions/{id} returns the full step timeline, including the initial user_input step.

Basic input/output (unary)

Before (legacy)

Python

# Request
interaction = client.interactions.create(
    model="gemini-3-flash-preview", input="Tell me a joke."
)

# Response access
print(interaction.outputs[0].text)

JavaScript

// Request
const interaction = await client.interactions.create({
    model: 'gemini-3-flash-preview',
    input: 'Tell me a joke.'
});

// Response access
console.log(interaction.outputs[0].text);

REST

// Request: POST /v1beta/interactions
{
  "model": "gemini-3-flash-preview",
  "input": "Tell me a joke."
}

// Response
{
  "id": "int_123",
  "role": "model",
  "outputs": [
    {
      "type": "text",
      "text": "Why did the chicken cross the road?"
    }
  ]
}

After (new schema)

Python

# Request
interaction = client.interactions.create(
    model="gemini-3-flash-preview", input="Tell me a joke."
)

# Response access
print(interaction.steps[-1].content[0].text)  # CHANGED: steps instead of outputs

JavaScript

// Request
const interaction = await client.interactions.create({
    model: 'gemini-3-flash-preview',
    input: 'Tell me a joke.'
});

// Response access
console.log(interaction.steps.at(-1).content[0].text);

REST

// Request: POST /v1beta/interactions
{
  "model": "gemini-3-flash-preview",
  "input": "Tell me a joke."
}

// POST Response
{
  "id": "int_123",
  "steps": [
    {
      "type": "model_output",
      "status": "done",
      "content": [
        {
          "type": "text",
          "text": "Why did the chicken cross the road?"
        }
      ]
    }
  ]
}

// GET /v1beta/interactions/int_123 (returns full timeline including input)
{
  "id": "int_123",
  "steps": [
    {
      "type": "user_input",
      "status": "done",
      "content": [
        { "type": "text", "text": "Tell me a joke." }
      ]
    },
    {
      "type": "model_output",
      "status": "done",
      "content": [
        {
          "type": "text",
          "text": "Why did the chicken cross the road?"
        }
      ]
    }
  ]
}

Function calling

The request structure remains unchanged, but the response replaces the flat outputs content with structured steps.

Before (legacy)

Python

# Accessing function call in legacy schema
for output in interaction.outputs:
    if output.type == "function_call":
        print(f"Calling {output.name} with {output.arguments}")

JavaScript

// Accessing function call in legacy schema
for (const output of interaction.outputs) {
    if (output.type === 'function_call') {
        console.log(`Calling {output.name} with {JSON.stringify(output.arguments)}`);
    }
}

REST

// Response
{
  "id": "int_001",
  "role": "model",
  "status": "requires_action",
  "outputs": [
    {
      "type": "thought",
      "signature": "abc123..."
    },
    {
      "type": "function_call",
      "id": "fc_1",
      "name": "get_weather",
      "arguments": { "location": "Boston, MA" }
    }
  ]
}

After (new schema)

Python

# Accessing function call in new steps schema
for step in interaction.steps:
    if step.type == "function_call":
        print(f"Calling {step.name} with {step.arguments}")

JavaScript

// Accessing function call in new steps schema
for (const step of interaction.steps) {
    if (step.type === 'function_call') {
        console.log(`Calling {step.name} with {JSON.stringify(step.arguments)}`);
    }
}

REST

// POST Response
{
  "id": "int_001",
  "status": "requires_action",
  "steps": [
    {
      "type": "thought",
      "status": "done",
      "signature": "abc123..."
    },
    {
      "type": "function_call",
      "status": "waiting",
      "id": "fc_1",
      "name": "get_weather",
      "arguments": { "location": "Boston, MA" }
    }
  ]
}

Server-side tools

Server-side tools (like Google Search or Code Execution) now yield specific step types in the steps array. While the legacy schema returned these operations as specific content types within the outputs array, the new schema moves them into the steps array. The following examples uses Google Search.

Before (legacy)

Python

# Accessing search results in legacy schema
for output in interaction.outputs:
    if output.type == "google_search_call":
        print(f"Searched for: {output.arguments.queries}")
    elif output.type == "google_search_result":
        print(f"Found results: {output.result.rendered_content}")

JavaScript

// Accessing search results in legacy schema
for (const output of interaction.outputs) {
    if (output.type === 'google_search_call') {
        console.log(`Searched for: {output.arguments.queries}`);
    } else if (output.type === 'google_search_result') {
        console.log(`Found results: {output.result.renderedContent}`);
    }
}

REST

// Request: POST /v1beta/interactions
{
  "model": "gemini-3-flash-preview",
  "input": "Who won the last Super Bowl?",
  "tools": [
    { "type": "google_search" }
  ]
}

// Response
{
  "id": "int_456",
  "outputs": [
    {
      "type": "google_search_call",
      "id": "gs_1",
      "arguments": { "queries": ["last Super Bowl winner"] }
    },
    {
      "type": "google_search_result",
      "call_id": "gs_1",
      "result": {
        "rendered_content": "<div>...</div>",
        "url": "https://www.nfl.com/super-bowl"
      }
    },
    {
      "type": "text",
      "text": "The Kansas City Chiefs won the last Super Bowl.",
      "annotations": [
        {
          "start_index": 4,
          "end_index": 22,
          "source": "https://www.nfl.com/super-bowl"
        }
      ]
    }
  ],
  "status": "completed"
}

After (new schema)

Python

# Accessing search results in new steps schema
for step in interaction.steps:
    if step.type == "google_search_call":
        print(f"Searched for: {step.arguments.queries}")
    elif step.type == "google_search_result":
        print(f"Found results: {step.result.search_suggestions}")

JavaScript

// Accessing search results in new steps schema
for (const step of interaction.steps) {
    if (step.type === 'google_search_call') {
        console.log(`Searched for: {step.arguments.queries}`);
    } else if (step.type === 'google_search_result') {
        console.log(`Found results: {step.result.searchSuggestions}`);
    }
}

REST

// Request: POST /v1beta/interactions
{
  "model": "gemini-3-flash-preview",
  "input": "Who won the last Super Bowl?",
  "tools": [
    { "type": "google_search" }
  ]
}

// POST Response
{
  "id": "int_456",
  "steps": [
    {
      "type": "google_search_call",
      "status": "done",
      "id": "gs_1",
      "arguments": { "queries": ["last Super Bowl winner"] },
      "signature": "abc123..."
    },
    {
      "type": "google_search_result",
      "status": "done",
      "call_id": "gs_1",
      "result": {
        "search_suggestions": "<div>...</div>"
      },
      "signature": "abc123..."
    },
    {
      "type": "model_output",
      "status": "done",
      "content": [
        {
          "type": "text",
          "text": "The Kansas City Chiefs won the last Super Bowl.",
          "annotations": [
            {
              "type": "url_citation",
              "url": "https://www.nfl.com/super-bowl",
              "title": "NFL.com",
              "start_index": 4,
              "end_index": 22
            }
          ]
        }
      ]
    }
  ],
  "status": "completed"
}

Streaming

Streaming exposes new event types:

New event types

  • interaction.created
  • interaction.status_update — now covers all lifecycle states including completion and errors (see statuses below)
  • step.start
  • step.delta
  • step.stop
interaction.status_update statuses
  • in_progress
  • active
  • completed
  • interrupted
  • requires_action
  • error

Deprecated event types

The following legacy event types are replaced by the new events listed above:

  • interaction.start → interaction.created
  • content.start → step.start
  • content.delta → step.delta
  • content.stop → step.stop
  • interaction.complete → interaction.status_update with status: "completed"
  • error → interaction.status_update with status: "error"
  • interaction.status_update → interaction.status_update (unchanged, but now covers additional states)

Streaming function calls: When you use streaming with function calling, the step.start event delivers the function name, and step.delta events stream the arguments as partial JSON strings (using arguments_delta). You must accumulate these deltas to get the full arguments. This differs from unary calls where you receive the complete function call object at once.

Examples

Before (Legacy)

Python

# Legacy streaming used content.delta
stream = client.interactions.create(
    model="gemini-3-flash-preview",
    input="Explain quantum entanglement in simple terms.",
    stream=True,
)

for chunk in stream:
    if chunk.event_type == "content.delta":
        if chunk.delta.type == "text":
            print(chunk.delta.text, end="", flush=True)

JavaScript

// Legacy streaming used content.delta
const stream = await client.interactions.create({
    model: 'gemini-3-flash-preview',
    input: 'Explain quantum entanglement in simple terms.',
    stream: true,
});

for await (const chunk of stream) {
    if (chunk.event_type === 'content.delta') {
        if (chunk.delta.type === 'text') {
            process.stdout.write(chunk.delta.text);
        }
    }
}

REST

// Request: POST /v1beta/interactions
{
  "model": "gemini-3-flash-preview",
  "input": "Explain quantum entanglement in simple terms.",
  "stream": true
}

// Response (SSE Lines)
// event: interaction.start
// data: {"id": "int_123", "status": "in_progress"}
//
// event: content.start
// data: {"index": 0, "type": "text"}
//
// event: content.delta
// data: {"delta": {"type": "text", "text": "Quantum entanglement is..."}}
//
// event: content.stop
// data: {"index": 0}
//
// event: interaction.complete
// data: {"id": "int_123", "status": "done", "usage": {"total_tokens": 42}}
After (New Schema)

Python

# Consuming stream and handling new event types
for event in client.interactions.create(
    model="gemini-3-flash-preview",
    input="Tell me a story.",
    stream=True,
):
    if event.type == "step.delta":  # CHANGED: step.delta instead of content.delta
        if event.delta.type == "text":
            print(event.delta.text, end="")

JavaScript

// Consuming stream and handling new event types
const stream = await client.interactions.create({
    model: 'gemini-3-flash-preview',
    input: 'Tell me a story.',
    stream: true,
});

for await (const event of stream) {
    if (event.type === 'step.delta') {  // CHANGED: step.delta instead of content.delta
        if (event.delta.type === 'text') {
            process.stdout.write(event.delta.text);
        }
    }
}

REST

 // Request: POST /v1beta/interactions
 // Accept: text/event-stream
 {
   "model": "gemini-3-flash-preview",
   "input": "Tell me a story."
 }

 // Response (SSE Lines)
 // event: interaction.created
 // data: {"type": "interaction.created", "interaction": {"id": "int_xyz", "status": "created"}} // CHANGED: 'type' instead of 'event_type'
 //
 // event: interaction.status_update
 // data: {"type": "interaction.status_update", "status": "in_progress"} // NEW: Lifecycle status updates in stream (postpone until Sessions launch dependency)
 //
 // event: step.start
 // data: {"type": "step.start", "index": 0, "step": {"type": "thought"}} // NEW: Replaces content.start, 'step' instead of 'content'
 //
 // event: step.delta
 // data: {"type": "step.delta", "index": 0, "delta": {"type": "thought", "text": "User wants an explanation."}} // NEW: Delta type matches step type
 //
 // event: step.stop
 // data: {"type": "step.stop", "index": 0, "status": "done"} // NEW: Includes status
 //
 // event: step.start
 // data: {"type": "step.start", "index": 1, "step": {"type": "model_output"}} // NEW: Step wrapper for output
 //
 // event: step.delta
 // data: {"type": "step.delta", "index": 1, "delta": {"type": "text", "text": "Hello"}}
 //
 // event: step.stop
 // data: {"type": "step.stop", "index": 1, "status": "done"}
 //
 // event: interaction.complete
 // data: {"type": "interaction.complete", "interaction": {"id": "int_xyz", "status": "completed", "usage": {"prompt_tokens": 10, "completion_tokens": 5, "total_tokens": 15}}} // NEW: End of stream event with interaction details

Stateless Conversation History

If you manage conversation history manually on the client side (stateless use case), you must update how you string along previous turns.

  • Legacy: Developers often collected the outputs array from responses and sent them back in the input field on the next turn.
  • New schema: You should now collect the steps array from the response and pass it in the input field of the next request, appending your new user turn as a user_input step.

Output format configuration: response_format changes

The updated API consolidates all output format controls into a unified, polymorphic response_format field. This centralizes output configuration at the top level and keeps generation_config focused on model behavior (like temperature, top_p, and thinking).

Key changes

  • The API removes response_mime_type. You now specify the MIME type per format entry inside response_format.
  • response_format is now a polymorphic object (or array). Each entry has a type discriminator (text, audio, image) and type-specific fields. To request multiple output modalities, pass an array of format entries.
  • image_config moves from generation_config to response_format. You now specify image output settings like aspect_ratio and image_size in a response_format entry with "type": "image".

Structured output (JSON)

The new schema removes the response_mime_type field. Instead, specify the MIME type and JSON schema inside a response_format object with "type": "text".

Before (legacy)

Python

interaction = client.interactions.create(
    model="gemini-3-flash-preview",
    input="Summarize this article.",
    response_mime_type="application/json",
    response_format={
        "type": "object",
        "properties": {
            "summary": {"type": "string"}
        }
    },
)

print(interaction.outputs[0].text)

JavaScript

const interaction = await client.interactions.create({
    model: 'gemini-3-flash-preview',
    input: 'Summarize this article.',
    responseMimeType: 'application/json',
    responseFormat: {
        type: 'object',
        properties: {
            summary: { type: 'string' }
        }
    },
});

console.log(interaction.outputs[0].text);

REST

// Request: POST /v1beta/interactions
{
  "model": "gemini-3-flash-preview",
  "input": "Summarize this article.",
  "response_mime_type": "application/json",
  "response_format": {
    "type": "object",
    "properties": {
      "summary": { "type": "string" }
    }
  }
}

After (new schema)

Python

interaction = client.interactions.create(
    model="gemini-3-flash-preview",
    input="Summarize this article.",
    # response_mime_type is removed — specify mime_type inside response_format
    response_format={
        "type": "text",
        "mime_type": "application/json",
        "schema": {
            "type": "object",
            "properties": {
                "summary": {"type": "string"}
            }
        }
    },
)

print(interaction.steps[-1].content[0].text)

JavaScript

const interaction = await client.interactions.create({
    model: 'gemini-3-flash-preview',
    input: 'Summarize this article.',
    // responseMimeType is removed — specify mimeType inside responseFormat
    responseFormat: {
        type: 'text',
        mimeType: 'application/json',
        schema: {
            type: 'object',
            properties: {
                summary: { type: 'string' }
            }
        }
    },
});

console.log(interaction.steps.at(-1).content[0].text);

REST

// Request: POST /v1beta/interactions
{
  "model": "gemini-3-flash-preview",
  "input": "Summarize this article.",
  // response_mime_type is removed
  "response_format": {
    "type": "text",                          // NEW: type discriminator
    "mime_type": "application/json",          // MOVED: from response_mime_type
    "schema": {                              // RENAMED: was response_format directly
      "type": "object",
      "properties": {
        "summary": { "type": "string" }
      }
    }
  }
}

Image configuration

The new schema removes image_config from generation_config. You now specify image output settings in a response_format entry with "type": "image".

Before (legacy)

Python

interaction = client.interactions.create(
    model="gemini-3-flash-preview",
    input="Generate an image of a sunset over the ocean.",
    generation_config={
        "image_config": {
            "aspect_ratio": "1:1",
            "image_size": "1K"
        }
    },
)

JavaScript

const interaction = await client.interactions.create({
    model: 'gemini-3-flash-preview',
    input: 'Generate an image of a sunset over the ocean.',
    generationConfig: {
        imageConfig: {
            aspectRatio: '1:1',
            imageSize: '1K'
        }
    },
});

REST

// Request: POST /v1beta/interactions
{
  "model": "gemini-3-flash-preview",
  "input": "Generate an image of a sunset over the ocean.",
  "generation_config": {
    "image_config": {
      "aspect_ratio": "1:1",
      "image_size": "1K"
    }
  }
}

After (new schema)

Python

interaction = client.interactions.create(
    model="gemini-3-flash-preview",
    input="Generate an image of a sunset over the ocean.",
    # image_config is removed from generation_config — use response_format
    response_format={
        "type": "image",
        "mime_type": "image/jpeg",
        "delivery": "inline",
        "aspect_ratio": "1:1",
        "image_size": "1K"
    },
)

JavaScript

const interaction = await client.interactions.create({
    model: 'gemini-3-flash-preview',
    input: 'Generate an image of a sunset over the ocean.',
    // imageConfig is removed from generationConfig — use responseFormat
    responseFormat: {
        type: 'image',
        mimeType: 'image/jpeg',
        delivery: 'inline',
        aspectRatio: '1:1',
        imageSize: '1K'
    },
});

REST

// Request: POST /v1beta/interactions
{
  "model": "gemini-3-flash-preview",
  "input": "Generate an image of a sunset over the ocean.",
  // image_config removed from generation_config
  "response_format": {
    "type": "image",                         // NEW: type discriminator
    "mime_type": "image/jpeg",
    "delivery": "inline",
    "aspect_ratio": "1:1",                   // MOVED: from generation_config.image_config
    "image_size": "1K"                       // MOVED: from generation_config.image_config
  }
}

To request multiple output modalities (for example, text and audio together), pass an array of format entries to response_format instead of a single object.

How to migrate to the new schema

SDK users

Upgrade to the latest SDK version (Python ≥1.76.0, JavaScript ≥1.53.0). The SDK automatically opts you into the new schema — no code changes needed beyond updating how you read responses (see examples above). Note that only the new schema is supported in these SDK versions. Older SDK versions (Python ≤1.73.1, JavaScript ≤1.50.1) will continue to work until the legacy schema is removed on June 6, 2026.

REST API users

Add the Api-Revision: 2026-05-20 header to your requests to opt in to the new schema now. After May 20, the new schema becomes the default for all requests. You can temporarily opt out with Api-Revision: 2026-05-06 until June 6, when the API permanently removes the legacy schema.

Timeline

Date Phase SDK users REST API users
May 6 Opt-in New major SDK version available (Python ≥2.0.0, JS ≥2.0.0). Upgrade to get the new schema automatically. Add Api-Revision: 2026-05-20 header to opt in. Default remains legacy.
May 20 Default flip No action needed if already upgraded. Older SDKs (Python 1.x.x, JS 1.x.x) still work but return legacy responses. New schema is now the default. Send Api-Revision: 2026-05-06 header to opt out.
June 6 Sunset 1.x.x SDK versions for Python and JS will break for Interactions API calls. Legacy schema removed for Interactions API. Api-Revision header ignored.

Migration Checklist

Steps schema (steps)

  • Update code to read response content from the steps array instead of outputs. See examples..
  • Verify that your code handles both user_input and model_output step types. See examples..
  • (Function Calling) Update code to find function_call steps in the steps array. See examples..
  • (Server-Side Tools) Update code to handle tool-specific steps (e.g., google_search_call, google_search_result). See examples.
  • (Stateless History) Update history management to pass the steps array in the input field of the next request. See details.
  • (Streaming only) Update client to listen for new SSE event types (interaction.created, step.delta, etc.). See examples.

Output format configuration (response_format)

  • Replace response_mime_type with a mime_type field inside response_format. See examples.
  • Wrap your existing response_format JSON schema inside a {"type": "text", "schema": ...} object. See examples.
  • (Image Generation) Move image_config from generation_config to a {"type": "image", ...} entry in response_format. See examples.
  • (Multimodal) Convert response_format from a single object to an array when requesting multiple output modalities..