OpenAI Responses API

ADK-Rust provides a dedicated client for OpenAI's Responses API (/v1/responses endpoint) — the successor to the Chat Completions API. The Responses API is the recommended way to use current GPT-5.6 models, including their full reasoning-effort range.

Overview

┌─────────────────────────────────────────────────────────────────────┐
│                  OpenAI Responses API Client                        │
├─────────────────────────────────────────────────────────────────────┤
│                                                                     │
│   Endpoint:  POST /v1/responses                                     │
│   Client:    OpenAIResponsesClient                                  │
│   Config:    OpenAIResponsesConfig                                  │
│   Feature:   openai                                                 │
│                                                                     │
│   Capabilities:                                                     │
│   • Streaming and non-streaming                                     │
│   • Reasoning summaries                                             │
│   • Tool / function calling                                         │
│   • Multi-turn via previous_response_id                             │
│   • Built-in tools (web search, file search, code interpreter)      │
│   • System instructions                                             │
│   • Model-aware sampling controls and max_output_tokens             │
│   • Automatic retry with exponential backoff                        │
│                                                                     │
│   vs Chat Completions (OpenAIClient):                               │
│   • Stateful conversations (server-side context)                    │
│   • Native reasoning summaries                                      │
│   • Built-in tool hosting                                           │
│   • Simpler multi-turn (no manual message history)                  │
│                                                                     │
└─────────────────────────────────────────────────────────────────────┘

When to Use Which Client

Use OpenAIResponsesClient when you need reasoning models with summaries, built-in tools, or want to use OpenAI's latest API. Use OpenAIClient for backward compatibility with existing Chat Completions workflows.

---

Installation

[dependencies]
adk-rust = { version = "2.1.0", features = ["openai"] }
adk-tool = "2.1.0"

Or with adk-model directly:

[dependencies]
adk-model = { version = "2.1.0", features = ["openai"] }

Set your API key:

export OPENAI_API_KEY="sk-..."

---

Quick Start

use adk_rust::prelude::*;
use adk_rust::session::{CreateRequest, SessionService};
use adk_rust::futures::StreamExt;
use adk_model::openai::{OpenAIResponsesClient, OpenAIResponsesConfig};
use std::collections::HashMap;
use std::sync::Arc;

#[tokio::main]
async fn main() -> anyhow::Result<()> {
    dotenvy::dotenv().ok();
    let api_key = std::env::var("OPENAI_API_KEY")?;

    // 1. Create the Responses API client
    let config = OpenAIResponsesConfig::new(&api_key, "gpt-5.6-terra");
    let model = Arc::new(OpenAIResponsesClient::new(config)?);

    // 2. Build an agent
    let agent = Arc::new(
        LlmAgentBuilder::new("assistant")
            .instruction("You are a helpful assistant. Be concise.")
            .model(model)
            .build()?,
    );

    // 3. Create a session
    let sessions: Arc<dyn SessionService> = Arc::new(InMemorySessionService::new());
    sessions.create(CreateRequest {
        app_name: "my_app".into(),
        user_id: "user".into(),
        session_id: Some("s1".into()),
        state: HashMap::new(),
    }).await?;

    // 4. Run through the Runner
    let runner = Runner::builder()
        .app_name("my_app")
        .agent(agent)
        .session_service(sessions)
        .build()?;

    let message = Content::new("user").with_text("What is the capital of France?");
    let mut stream = runner.run(
        adk_rust::UserId::new("user")?,
        adk_rust::SessionId::new("s1")?,
        message,
    ).await?;

    while let Some(event) = stream.next().await {
        let event = event?;
        if let Some(content) = &event.llm_response.content {
            for part in &content.parts {
                if let Some(text) = part.text() {
                    print!("{text}");
                }
            }
        }
    }
    println!();
    Ok(())
}

---

Configuration

Basic Configuration

use adk_model::openai::OpenAIResponsesConfig;

// Minimal — just API key and model
let config = OpenAIResponsesConfig::new("sk-...", "gpt-5.6-luna");

// With organization and project
let config = OpenAIResponsesConfig::new("sk-...", "gpt-5.6-terra")
    .with_organization("org-...")
    .with_project("proj-...");

// Custom base URL (for proxies or compatible APIs)
let config = OpenAIResponsesConfig::new("sk-...", "gpt-5.6-terra")
    .with_base_url("https://my-proxy.example.com/v1");

Reasoning Models

For GPT-5.6 reasoning models, configure reasoning effort and summary:

use adk_model::openai::{
    OpenAIReasoningEffort, OpenAIResponsesClient,
    OpenAIResponsesConfig, ReasoningSummary,
};

let config = OpenAIResponsesConfig::new("sk-...", "gpt-5.6-terra")
    .with_reasoning_summary(ReasoningSummary::Detailed);

let model = OpenAIResponsesClient::new_with_reasoning_effort(
    config,
    OpenAIReasoningEffort::Max,
)?;

GPT-5.6 supports None, Low, Medium, High, XHigh, and Max through the Responses API. Chat Completions supports up to XHigh.

Reasoning summaries appear as Part::Thinking in the response stream, letting you show the model's thought process to users.

Retry Configuration

use adk_model::retry::RetryConfig;

let client = OpenAIResponsesClient::new(config)?
    .with_retry_config(RetryConfig {
        max_retries: 3,
        ..Default::default()
    });

Retries are automatic for rate limits (429), server errors (500/502/503/504), and network failures.

---

Available Models

---

Features

Tool Calling

Function tools work the same way as with OpenAIClient — define tools on the agent and the runner handles the tool call loop:

use adk_rust::prelude::*;
use adk_model::openai::{OpenAIResponsesClient, OpenAIResponsesConfig};
use adk_tool::FunctionTool;
use std::sync::Arc;

async fn get_weather(
    _ctx: Arc<dyn ToolContext>,
    args: serde_json::Value,
) -> Result<serde_json::Value> {
    let city = args["city"].as_str().unwrap_or("unknown");
    Ok(serde_json::json!({
        "city": city,
        "temperature_f": 72,
        "conditions": "Sunny"
    }))
}

let weather_tool = FunctionTool::new(
    "get_weather",
    "Get current weather for a city. Requires a 'city' string parameter.",
    get_weather,
);

let config = OpenAIResponsesConfig::new(&api_key, "gpt-5.6-terra");
let model = Arc::new(OpenAIResponsesClient::new(config)?);

let agent = LlmAgentBuilder::new("weather_agent")
    .instruction("Use the get_weather tool to answer weather questions.")
    .model(model)
    .tool(Arc::new(weather_tool))
    .build()?;

Multi-Turn Conversations

The Runner automatically manages conversation history through sessions. Each turn's context is preserved:

// Turn 1
let msg1 = Content::new("user").with_text("My name is Alice.");
let mut stream = runner.run(uid.clone(), sid.clone(), msg1).await?;
// ... consume stream ...

// Turn 2 — the model remembers the previous turn
let msg2 = Content::new("user").with_text("What is my name?");
let mut stream = runner.run(uid.clone(), sid.clone(), msg2).await?;
// Response: "Your name is Alice."

Per-Request Reasoning Override

Override reasoning settings per-request using LlmRequest extensions:

use adk_rust::prelude::*;

let agent = LlmAgentBuilder::new("flexible_reasoner")
    .model(model)
    .generate_content_config(GenerateContentConfig {
        extensions: {
            let mut ext = std::collections::HashMap::new();
            ext.insert("openai".to_string(), serde_json::json!({
                "reasoning": {
                    "effort": "high",
                    "summary": "detailed"
                }
            }));
            ext
        },
        ..Default::default()
    })
    .build()?;

Built-In Tools

The Responses API supports OpenAI-hosted tools. Prefer the typed wrappers from adk-tool:

use adk_tool::OpenAIWebSearchTool;
use std::sync::Arc;

let agent = LlmAgentBuilder::new("researcher")
    .model(model)
    .tool(Arc::new(OpenAIWebSearchTool::new().preview()))
    .build()?;

Available wrappers include OpenAIWebSearchTool, OpenAIFileSearchTool, OpenAICodeInterpreterTool, OpenAIImageGenerationTool, OpenAIComputerUseTool, OpenAIMcpTool, OpenAILocalShellTool, OpenAIShellTool, and OpenAIApplyPatchTool.

Previous Response ID

For server-side conversation state (bypassing local session history), pass previous_response_id:

let agent = LlmAgentBuilder::new("stateful")
    .model(model)
    .generate_content_config(GenerateContentConfig {
        extensions: {
            let mut ext = std::collections::HashMap::new();
            ext.insert("openai".to_string(), serde_json::json!({
                "previous_response_id": "resp_abc123"
            }));
            ext
        },
        ..Default::default()
    })
    .build()?;

---

Streaming Behavior

The Responses API client streams text and reasoning deltas in real-time:

• Text deltas arrive as Part::Text with partial: true
• Reasoning summary deltas arrive as Part::Thinking with partial: true
• Function calls are emitted from the final ResponseCompleted event with correct names and arguments
• The final event has turn_complete: true with usage metadata and finish reason

This means you see text appearing token-by-token while the model generates, and function calls arrive as complete objects ready for execution.

---

Provider Metadata

Every response includes provider metadata with the response_id:

if let Some(meta) = &response.provider_metadata {
    let response_id = meta["openai"]["response_id"].as_str();
    // Use for previous_response_id, logging, debugging
}

Additional metadata may include:

• encrypted_content — from reasoning models (for context preservation)
• built_in_tool_outputs — results from web search, file search, code interpreter

---

Error Handling

Errors are mapped to structured AdkError with appropriate categories:

match runner.run(uid, sid, message).await {
    Ok(stream) => { /* process stream */ }
    Err(e) if e.is_retryable() => { /* retry logic */ }
    Err(e) if e.is_unauthorized() => { /* check API key */ }
    Err(e) => { /* handle other errors */ }
}

---

Background Mode & Cancellation

For long-running requests, submit with background: true and poll for completion:

use adk_model::openai::{OpenAIResponsesClient, OpenAIResponsesConfig};

let client = OpenAIResponsesClient::new(config)?;

// Submit with background: true via extensions
let mut gen_config = GenerateContentConfig::default();
gen_config.extensions.insert("openai".into(), serde_json::json!({ "background": true }));

// ... send request, extract response_id from provider_metadata ...

// Poll until terminal status
let response = client.poll_response("resp_abc123").await?;
// Check provider_metadata["openai"]["status"]: "completed", "in_progress", "failed", "cancelled"

// Cancel a running background response
let cancelled = client.cancel_response("resp_abc123").await?;

Deep research models (o3-deep-research, o4-mini-deep-research) automatically enable background mode without explicit background: true.

---

Example

A complete 7-scenario example is available at examples/openai_responses/:

export OPENAI_API_KEY=sk-...
cargo run --manifest-path examples/openai_responses/Cargo.toml

Scenarios covered:

1. Basic non-streaming chat
2. Basic streaming chat
3. Reasoning model with summary (o4-mini compatibility path)
4. Tool calling with function tools
5. Multi-turn conversation
6. System instructions
7. Temperature and generation config (gpt-4.1-nano compatibility path)

Additional Examples

Six standalone example crates demonstrate specific Responses API features:

---

Related

• Cloud Model Providers — All supported LLM providers
• Ollama (Local) — Run models locally
• LlmAgent — Using models with agents
• Function Tools — Adding tools to agents

---

Previous: ← Cloud Providers | Next: Ollama (Local) →