Open Responses: What you need to know

Summary

Open Responses is a new open inference standard initiated by OpenAI, developed by the open source AI community, and backed by Hugging Face. While agentic workloads require multi-step planning and tool execution, the ecosystem has remained constrained by the legacy Chat Completion format and proprietary Responses implementations. Open Responses addresses this mismatch by standardizing semantic event streaming, provider-specific configuration options, and distinct roles for model providers and intermediary routers. The specification formalizes reasoning visibility across raw traces, protected encrypted content, and sanitized summaries while supporting provider-managed sub-agent execution loops. Developers can now orchestrate multi-step tool calls and stream reasoning deltas across heterogeneous providers through a single consistent API schema.

Context

Autonomous AI agents require systems that reason, plan, and act over long time horizons, but much of the inference ecosystem still relies on the legacy Chat Completion format designed for turn-based conversation. Although the Responses API format was designed to solve these agentic limitations, its closed nature hindered broad adoption and cross-provider standardization.

Approach / What changed

Open Responses open-sources and extends the Responses API into a shared standard featuring stateless operations, semantic event streaming, standardized model configurations, and separated definitions for Model Providers and Routers. It formalizes raw reasoning, encrypted reasoning, and summary streams, while natively handling internally and externally executed tool-calling loops on the provider side.

Takeaways

• Open Responses formalizes three reasoning item fields: content for raw reasoning traces, encrypted_content for protected content, and summary for sanitized traces.
• The standard natively supports both internally hosted tools executed entirely within the provider infrastructure and externally hosted tools like client functions or MCP servers.
• Provider-managed sub-agent loops enable multi-step reasoning and tool execution cycles in a single request, controlled by max_tool_calls and tool_choice parameters.

Related reading

• Docker · Jul 16, 2026

  AI Agents Explained: How to Build with Them Safely

  AI agents combine reasoning models, external tools, memory, orchestration, and execution environments to autonomously pursue goals across iterative perceive-reason-act loops. While simple chatbots only generate text responses to direct prompts, autonomous agents execute multi-step operations such as querying databases, modifying codebases, and executing shell commands without requiring continuous manual approvals. However, this unchecked operational autonomy significantly widens the blast radius when models encounter logic errors or vague instructions, risking destructive filesystem operations and credential leaks. Safe deployment requires shifting focus from model-level guardrails to infrastructure-level containment strategies, such as scoping tool access and isolating execution inside disposable microVM sandboxes. Controlling networking boundaries, file access, and credential exposure ensures teams achieve automated execution velocity without exposing developer host environments to uncontained systemic risk.

  Jin Kim

• Github · Aug 10, 2026