home.social

#responsesapi — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #responsesapi, aggregated by home.social.

  1. OpenAI integriert WebSockets in die Responses API und senkt die Latenz bei Agenten-Workflows um bis zu 40 Prozent.

    Durch persistente Verbindungen und serverseitiges Kontext-Caching entfallen redundante HTTP-Anfragen. Modelle wie GPT-5.3-Codex generieren so auf Cerebras-Hardware bis zu 4.000 Token pro Sekunde. Drittanbieter verzeichnen messbare Leistungssprünge.

    #OpenAI #ResponsesAPI #Cerebras #LLM #News
    all-ai.de/news/news26/openai-a

  2. OpenAI integriert WebSockets in die Responses API und senkt die Latenz bei Agenten-Workflows um bis zu 40 Prozent.

    Durch persistente Verbindungen und serverseitiges Kontext-Caching entfallen redundante HTTP-Anfragen. Modelle wie GPT-5.3-Codex generieren so auf Cerebras-Hardware bis zu 4.000 Token pro Sekunde. Drittanbieter verzeichnen messbare Leistungssprünge.

    #OpenAI #ResponsesAPI #Cerebras #LLM #News
    all-ai.de/news/news26/openai-a

  3. OpenAI integriert WebSockets in die Responses API und senkt die Latenz bei Agenten-Workflows um bis zu 40 Prozent.

    Durch persistente Verbindungen und serverseitiges Kontext-Caching entfallen redundante HTTP-Anfragen. Modelle wie GPT-5.3-Codex generieren so auf Cerebras-Hardware bis zu 4.000 Token pro Sekunde. Drittanbieter verzeichnen messbare Leistungssprünge.

    #OpenAI #ResponsesAPI #Cerebras #LLM #News
    all-ai.de/news/news26/openai-a

  4. OpenAI integriert WebSockets in die Responses API und senkt die Latenz bei Agenten-Workflows um bis zu 40 Prozent.

    Durch persistente Verbindungen und serverseitiges Kontext-Caching entfallen redundante HTTP-Anfragen. Modelle wie GPT-5.3-Codex generieren so auf Cerebras-Hardware bis zu 4.000 Token pro Sekunde. Drittanbieter verzeichnen messbare Leistungssprünge.

    #OpenAI #ResponsesAPI #Cerebras #LLM #News
    all-ai.de/news/news26/openai-a

  5. OpenAI integriert WebSockets in die Responses API und senkt die Latenz bei Agenten-Workflows um bis zu 40 Prozent.

    Durch persistente Verbindungen und serverseitiges Kontext-Caching entfallen redundante HTTP-Anfragen. Modelle wie GPT-5.3-Codex generieren so auf Cerebras-Hardware bis zu 4.000 Token pro Sekunde. Drittanbieter verzeichnen messbare Leistungssprünge.

    #OpenAI #ResponsesAPI #Cerebras #LLM #News
    all-ai.de/news/news26/openai-a

  6. ----------------

    🎯 AI
    ===================

    Executive summary: OpenAI introduces three agentic primitives to make long-horizon, knowledge-work agents practical: reusable Skills (with a SKILL.md manifest), an upgraded Shell execution environment (hosted containers accessible via the Responses API), and server-side compaction for maintaining long conversational context.

    Technical details:
    • Skills are versioned bundles of files plus a SKILL.md manifest exposing metadata (name, description, path) that the model uses to decide invocation.
    • The Shell tool provides a real terminal environment implemented as OpenAI-hosted containers (or local shells with identical semantics) that support stateful runs, tool calls, multi-turn continuations, and artifact creation.
    • Compaction operates server-side to compress conversation history when context thresholds are crossed; two modes exist: automatic in-stream compaction and an explicit /responses/compact endpoint for manual control.

    How it works conceptually:
    • Routing: the model reads skill metadata (name, description, path) and only loads SKILL.md when it decides to invoke the skill, reducing prompt bloat.
    • Execution: hosted Shell runs through the Responses API, coupling tool execution and artifact storage with the agent’s conversation state.
    • Continuity: compaction preserves semantic continuity by reducing token footprint while retaining essential context for long workflows.

    Best practices and patterns observed:
    • Skill descriptions should be written as routing logic: concrete “Use when” vs. “Don’t use when,” explicit outputs, and success criteria.
    • Include negative examples and edge cases in skill metadata to reduce misfires; Glean reported an initial ~20% drop in correct triggers that recovered after adding negative examples.
    • Keep stable procedures in Skills (versioned playbooks) and leave dynamic planning in the model to avoid prompt spaghetti.

    Limitations and considerations:
    • Skills require careful authoring of routing metadata to avoid misrouting.
    • Hosted Shell access implies controlled internet/process boundaries; security and audit controls around hosted execution environments are critical in deployment designs.

    🔹 agents #skills #compaction #ResponsesAPI

    🔗 Source: developers.openai.com/blog/ski

  7. 🛡️ Enterprise-grade security with proper authentication handling, structured responses containing tool discovery, reasoning steps, and execution results

    💰 Simple pricing model - pay only for tokens consumed by selected #GroqCloud model, bring your own #MCP server and API key with third-party fees billed directly

    🔗 Works with #ResponsesAPI (native #MCP support) and #ChatCompletions API (retrofitted). #ResponsesAPI recommended for multi-step workflows and approval controls

  8. 🛡️ Enterprise-grade security with proper authentication handling, structured responses containing tool discovery, reasoning steps, and execution results

    💰 Simple pricing model - pay only for tokens consumed by selected #GroqCloud model, bring your own #MCP server and API key with third-party fees billed directly

    🔗 Works with #ResponsesAPI (native #MCP support) and #ChatCompletions API (retrofitted). #ResponsesAPI recommended for multi-step workflows and approval controls

  9. 🛡️ Enterprise-grade security with proper authentication handling, structured responses containing tool discovery, reasoning steps, and execution results

    💰 Simple pricing model - pay only for tokens consumed by selected #GroqCloud model, bring your own #MCP server and API key with third-party fees billed directly

    🔗 Works with #ResponsesAPI (native #MCP support) and #ChatCompletions API (retrofitted). #ResponsesAPI recommended for multi-step workflows and approval controls

  10. 🛡️ Enterprise-grade security with proper authentication handling, structured responses containing tool discovery, reasoning steps, and execution results

    💰 Simple pricing model - pay only for tokens consumed by selected #GroqCloud model, bring your own #MCP server and API key with third-party fees billed directly

    🔗 Works with #ResponsesAPI (native #MCP support) and #ChatCompletions API (retrofitted). #ResponsesAPI recommended for multi-step workflows and approval controls

  11. 🛡️ Enterprise-grade security with proper authentication handling, structured responses containing tool discovery, reasoning steps, and execution results

    💰 Simple pricing model - pay only for tokens consumed by selected #GroqCloud model, bring your own #MCP server and API key with third-party fees billed directly

    🔗 Works with #ResponsesAPI (native #MCP support) and #ChatCompletions API (retrofitted). #ResponsesAPI recommended for multi-step workflows and approval controls

  12. Oh, the brave new world of the Responses API, where the most groundbreaking feature is to, wait for it... enable #JavaScript and cookies! 🍪💻 Because clearly, that's something revolutionary in 2023. 🚀🎉
    openai.com/index/new-tools-and #ResponsesAPI #Cookies #Innovation #TechNews #HackerNews #ngated

  13. Oh, the brave new world of the Responses API, where the most groundbreaking feature is to, wait for it... enable #JavaScript and cookies! 🍪💻 Because clearly, that's something revolutionary in 2023. 🚀🎉
    openai.com/index/new-tools-and #ResponsesAPI #Cookies #Innovation #TechNews #HackerNews #ngated

  14. Oh, the brave new world of the Responses API, where the most groundbreaking feature is to, wait for it... enable #JavaScript and cookies! 🍪💻 Because clearly, that's something revolutionary in 2023. 🚀🎉
    openai.com/index/new-tools-and #ResponsesAPI #Cookies #Innovation #TechNews #HackerNews #ngated

  15. Oh, the brave new world of the Responses API, where the most groundbreaking feature is to, wait for it... enable #JavaScript and cookies! 🍪💻 Because clearly, that's something revolutionary in 2023. 🚀🎉
    openai.com/index/new-tools-and #ResponsesAPI #Cookies #Innovation #TechNews #HackerNews #ngated

  16. #OpenAI Launches New Tools for Building #AI Agents with #ResponsesAPI, Web Search, File Search & Computer Use Features 🤖
    Today, #OpenAI introduces new #APIs and tools designed specifically to simplify development of agentic applications:

  17. • 🔍 New #ResponsesAPI combining simplicity of Chat Completions with tool-use capabilities for building agents
    • 🌐 Built-in web search tool providing fast, up-to-date answers with clear citations (available with #GPT4o and GPT-4o-mini)

  18. #OpenAI Launches New Tools for Building #AI Agents with #ResponsesAPI, Web Search, File Search & Computer Use Features 🤖
    Today, #OpenAI introduces new #APIs and tools designed specifically to simplify development of agentic applications:

  19. • 🔍 New #ResponsesAPI combining simplicity of Chat Completions with tool-use capabilities for building agents
    • 🌐 Built-in web search tool providing fast, up-to-date answers with clear citations (available with #GPT4o and GPT-4o-mini)

  20. • 🔍 New #ResponsesAPI combining simplicity of Chat Completions with tool-use capabilities for building agents
    • 🌐 Built-in web search tool providing fast, up-to-date answers with clear citations (available with #GPT4o and GPT-4o-mini)

  21. #OpenAI Launches New Tools for Building #AI Agents with #ResponsesAPI, Web Search, File Search & Computer Use Features 🤖
    Today, #OpenAI introduces new #APIs and tools designed specifically to simplify development of agentic applications:

  22. • 🔍 New #ResponsesAPI combining simplicity of Chat Completions with tool-use capabilities for building agents
    • 🌐 Built-in web search tool providing fast, up-to-date answers with clear citations (available with #GPT4o and GPT-4o-mini)

  23. #OpenAI Launches New Tools for Building #AI Agents with #ResponsesAPI, Web Search, File Search & Computer Use Features 🤖
    Today, #OpenAI introduces new #APIs and tools designed specifically to simplify development of agentic applications:

  24. #OpenAI Launches New Tools for Building #AI Agents with #ResponsesAPI, Web Search, File Search & Computer Use Features 🤖
    Today, #OpenAI introduces new #APIs and tools designed specifically to simplify development of agentic applications:

  25. • 🔍 New #ResponsesAPI combining simplicity of Chat Completions with tool-use capabilities for building agents
    • 🌐 Built-in web search tool providing fast, up-to-date answers with clear citations (available with #GPT4o and GPT-4o-mini)

  26. #OpenAI Launches New Tools for Building #AI #Agents with #ResponsesAPI, Web Search, File Search & Computer Use Features 🤖

    OpenAI releases its first set of building blocks to help #developers build useful and reliable agents—systems that independently accomplish tasks on behalf of users. 🧵👇