Responses API
TrialPlatforms
OpenAI’s API format for generating text, structured outputs, media, and agentic loops.
Why it's here
Placed in Trial: 14 article(s) of evidence from 5 source(s), led by framework updates, with 1 in the last 30 days. Confidence 90%.
Evidence (14)
- 8The New Stack·10/7/2026model_releaseGPT-6.1 Sol matches Astra’s accuracy at much lower cost
OpenAI launched GPT-6.1 Sol as a cheaper upgrade to GPT-6 Sol and a near-match for GPT-6 Astra. In benchmark-style tests covering CI triage, incident analysis, and code generation, GPT-6.1 Sol matched Astra’s accuracy while using less time, fewer tokens in some cases, and significantly less cost.
- 6Simon Willison·8/22/2026framework_updatellm 0.33 adds embedding key support and template chaining
llm 0.33 updates the project to OpenAI Python library 3.x and switches its HTTP client dependency from httpx to httpx2. It also adds per-call --key support for embedding commands and allows repeating llm prompt templates to combine configuration and prompts, while expanding Responses API support with reasoning_summary options.
- 4Simon Willison·8/21/2026framework_updatellm-openrouter 0.7 adds Responses API support and new tools
Simon Willison released llm-openrouter 0.7, updating the plugin for compatibility with LLM 0.32. The update switches models to OpenRouter's Responses API implementation and adds three server-side tools: Shell, WebFetch, and WebSearch.
- 6The New Stack·8/21/2026model_releaseDeepSeek releases v4 Flash Vision experimental model
DeepSeek has introduced deepseek-v4-flash-vision-exp, a vision-capable model that can accept images together with text. The documentation explains supported formats, input methods, and request-size limits for inline uploads, external URLs, and Files API references.
- 4Hacker News·8/21/2026product_launchAn Argument for Replacing TUIs with Native Apps
The post argues that terminal user interfaces are overused and that many tasks are better served by native graphical applications. It showcases several personally built apps, including a Markdown viewer, a SageMath calculator-style frontend, an AI-assisted music player, and a self-driving wiki, all largely generated with modern tools and agents. The core message is that native UI development can be delegated more than many developers assume.
- 4Simon Willison·8/15/2026product_launchCORS Chat web UI for testing OpenAI-compatible endpoints
Simon Willison built CORS Chat, a web UI for exercising OpenAI Responses-compatible chat endpoints while testing models running locally and through hosted providers. It was used with LM Studio's CORS support and OpenRouter, and it can persist conversations in the browser, export them as JSON, and progressively render SVG images as they stream in.
- 7Hacker News·8/14/2026model_releaseDeepSeek launches V4-Pro and adds peak/off-peak API pricing
DeepSeek has launched DeepSeek-V4-Pro with upgraded agent capabilities, flexible reasoning effort levels, and native support for the OpenAI Responses API. Alongside the release, the company is updating API pricing to include peak and off-peak rates, with off-peak usage set at 50% lower cost starting Aug. 16, 2026.
- 5OpenAI Blog·8/13/2026model_releaseOpenAI releases builder guide for GPT-5.6
OpenAI published a builder-focused guide explaining how startups can use GPT-5.6 to build AI agents faster and at lower cost. The guide highlights smarter model selection and new capabilities in the Responses API for developer workflows.
- 8Simon Willison·8/4/2026framework_updateLLM 0.32 adds reasoning traces, OpenAI Responses support, and smarter logging
LLM 0.32 is a major release that adds visible reasoning traces, support for server-side tools, redesigned content-addressable SQLite logs, and new model support including GPT-5.6. The update also introduces Python API changes for message-based prompts and event streaming, plus a new llm-chat-completions-server plugin and substantial llm-anthropic updates.
- 8Simon Willison·7/31/2026model_releaseDeepSeek V4-Flash API Enters Public Beta
DeepSeek has released the V4-Flash API in public beta, with the calling interface unchanged but the model name updated to deepseek-v4-flash. The update claims stronger agent performance, native support for the Responses API format, and adaptation for Codex, while the V4-Pro API and app/web models remain unchanged.
- 6OpenAI Blog·4/22/2026framework_updateWebSockets speed up Codex agent workflows in Responses API
OpenAI details how its Codex agent loop uses WebSockets and connection-scoped caching to reduce API overhead. The changes improve model latency and make agentic workflows more responsive.
- 7OpenAI Blog·3/11/2026framework_updateOpenAI adds secure computer environments to Responses API
OpenAI says it has built an agent runtime on top of the Responses API using a shell tool and hosted containers. The setup is designed to let agents use files, tools, and persistent state in a secure, scalable environment.
- 6OpenAI Blog·1/23/2026framework_updateInside the Codex Agent Loop
OpenAI provides a technical deep dive into how the Codex agent loop works in Codex CLI. The post explains how it orchestrates models, tools, prompts, and performance using the Responses API.
- 7Hugging Face Blog·1/15/2026open_sourceHugging Face backs Open Responses for agent workflows
Hugging Face introduced Open Responses, an open inference standard based on OpenAI’s Responses API and intended for agentic workloads. The post explains how it aims to replace Chat Completions with a more flexible, open format for stateless requests, semantic streaming, and provider-specific extensions.