Responses API
Responses API is a server-side API product from OpenAI designed for building and managing AI model conversational interaction workflows. Key features include server-side maintenance of conversation history to reduce client-side context management overhead, support for integrated tool calls and file retrieval capabilities, and a unified interface for invoking multiple OpenAI language models. It is intended for developers building complex AI applications.
Core Facts
Timeline (last 90 days)
The Responses API adopts a structured event stream, pushing semantic events such as response.output_item.added, response.output_text.delta, and response.output_item.done, and uses fields such as output_index and item_id to identify the position of each output item in the final response.
Consumers using the Responses API must maintain the output_index → block mapping themselves.
The Responses API allows conversation state to be saved server-side, so the client only needs to pass previous_response_id to continue the context, and attaches built-in tools as first-class citizens directly to the request.
On Chat Completions, gpt-6-luna's function calling is available only when reasoning effort is none; using built-in tools requires switching to the Responses API.
The Responses API on most LLM gateways only supports GPT-series models.
OpenAI officially claims the Decisions API is about 10 times faster than the Responses API, but it did not disclose benchmarks and did not publish head-to-head accuracy comparisons with competitors.
OpenAI claims the Decisions API is 10 times faster than the Responses API, but attaches no benchmark data; Microsoft and other vendors have not published latency data.
Official data shows that the Decisions API returns results approximately 10 times faster than the Responses API
Traditional large model APIs (such as the Responses API) typically return free-form text, requiring developers to add an additional parsing layer to extract structured information
The Responses API is the most low-level path among OpenAI's three construction approaches, with its core composition being the model plus managed tools and custom tools, giving developers maximum control
17 more timeline events
All Facts (20)
xAI provides an OpenAI-compatible API interface, which was the basis for early @ai-sdk/xai being able to adapt via the Chat Completions endpoint.
90%VerifiedOpenAI于2024年推出了Responses API，引入了内置工具调用和状态管理能力
80%UnverifiedResponses API引入了previous_response_id参数实现多轮对话的自动上下文串联
85%UnverifiedThe Responses API is an evolved version of the Chat Completions API, integrating built-in tools such as web search, file search, and a code interpreter.
70%UnverifiedThe Responses API natively supports a variety of built-in tools, including web search, file search, code interpreter, and computer use.
60%UnverifiedThe Responses API is a next-generation API interface launched by OpenAI in 2025, intended to replace the Chat Completions API.
60%UnverifiedConsumers using the Responses API must maintain the output_index → block mapping themselves.
50%UnverifiedThe Responses API adopts a structured event stream, pushing semantic events such as response.output_item.added, response.output_text.delta, and response.output_item.done, and uses fields such as output_index and item_id to identify the position of each output item in the final response.
50%UnverifiedThe Responses API allows conversation state to be saved server-side, so the client only needs to pass previous_response_id to continue the context, and attaches built-in tools as first-class citizens directly to the request.
50%UnverifiedOn Chat Completions, gpt-6-luna's function calling is available only when reasoning effort is none; using built-in tools requires switching to the Responses API.
50%UnverifiedThe Responses API on most LLM gateways only supports GPT-series models.
50%UnverifiedOpenAI officially claims the Decisions API is about 10 times faster than the Responses API, but it did not disclose benchmarks and did not publish head-to-head accuracy comparisons with competitors.
50%UnverifiedOpenAI claims the Decisions API is 10 times faster than the Responses API, but attaches no benchmark data; Microsoft and other vendors have not published latency data.
50%UnverifiedTraditional large model APIs (such as the Responses API) typically return free-form text, requiring developers to add an additional parsing layer to extract structured information
50%UnverifiedOfficial data shows that the Decisions API returns results approximately 10 times faster than the Responses API
50%UnverifiedThe Responses API is the most low-level path among OpenAI's three construction approaches, with its core composition being the model plus managed tools and custom tools, giving developers maximum control
50%UnverifiedBoth Anthropic's Claude and OpenAI's Responses API implement the MCP client standard, and therefore can use the same Wassenger MCP server.
50%UnverifiedIntegrating the OpenAI Responses API with MCP only requires adding one item to the tools array, setting type to mcp and attaching a label, server address, and key.
50%UnverifiedThe none effort parameter has been removed from the schema, and legacy system tool-calling capabilities that relied on this parameter to execute function calls at the chat completions endpoint will stop working and must be migrated to the Responses API.
50%UnverifiedProjects still relying on the legacy completions endpoint should migrate to the Chat Completions or Responses API as soon as possible to avoid future compatibility issues.
50%