Skip to content

OpenAI Ties Reasoning Items to Conversation History in the Responses API

OpenAI Developers · XFrom the source· @OpenAIDevs·· 1 day agoPicks

22.3K views · 287 likes · 17 reposts (at collection)

Original

What happened

OpenAI says agents using GPT-6.2 in the Responses API can build on earlier reasoning to deliver better results with fewer tokens and more cache hits. Starting with its newest frontier models, each reasoning item is tied to the preceding conversation history, so editing earlier messages causes the API to drop later reasoning.

Why it matters

When earlier messages are edited, the Responses API drops subsequent reasoning items tied to the previous conversation history, while agents using GPT-6.2 can build on earlier reasoning with fewer tokens and more cache hits.

Article

Agents using GPT-6.2 in the Responses API can build on earlier reasoning to deliver better results with fewer tokens and more cache hits.

Starting with our newest frontier models, each reasoning item is now tied to the conversation history before it, so editing earlier messages causes the API to drop reasoning that follows the change.

2 reports

    The story so far

    OpenAI released GPT-6.2 Astra and GPT-6.2 Sol in ChatGPT Work, Codex, and the API, with Astra priced 20% lower per input and output token than GPT-6 Astra and Sol keeping GPT-6.1 Sol's token pricing. OpenAI also expanded Ultrafast mode to Pro 500 subscribers for both models and to Pro 200 subscribers starting with Sol, while cutting Ultrafast API and Enterprise pricing by 50% for three months. Because of increased capabilities, OpenAI launched both models with tighter cybersecurity safeguards that it said may increase refusals on some security research tasks, and advised developers using the Responses API to append new turns rather than edit earlier messages, as editing drops later reasoning tied to the conversation history. OpenAI reported new highs for Astra on Agents' Last Exam, BenchCAD 2.1 Core, and Terminal-Bench Science 0.1, while Greg Brockman said Astra exceeded the human baseline on a physical 3D understanding benchmark, and Perplexity reported higher scores on its WANDR benchmark after adding both models to Perplexity Search and Computer. Independent feedback has been mixed: Epoch AI reported that Astra scored 65% on its Epoch Automation Reports benchmark, trailing Fable 5.2's 82%, and developer Simon Willison said neither model handled his animation benchmark well compared with Claude Fable 5.2 and Gemini 4 Carbon. Solveit's Jeremy Howard also said he reverted Astra and Sol to pre-6.2 versions because the update was not working for him or his users.

    How the story developed · 5

    View full event
    1. 11/20 08:00 (UTC+8)

      OpenAI Ties Reasoning Items to Conversation History in the Responses API This report

    2. 11/19 00:33 (UTC+8)
      Greg Brockman Says GPT-6.2 Is Incredible at Browser UseGreg Brockman · X
    3. 11/18 06:15 (UTC+8)
      Tibor Blaho Tests GPT-6.2 Astra and Sol on Daily WorkTibor Blaho · X
    4. 11/18 06:15 (UTC+8)
      Perplexity Shares GPT-6.2 Astra and Sol WANDR EvalsAravind Srinivas · X
    5. 11/18 02:36 (UTC+8)
      OpenAI Introduces GPT-6.2 Astra and SolOpenAI News

    Read next

    See the full storyOpenAI Launches GPT-6.2 Astra and Sol, Drawing Mixed Reviews and Some Version Reversions
    This story is in 11/21 AI DailyRead the whole issue