Keep getting "Item with id 'rs_xxxxx' not found. Items are not persisted when 'store' is set to false" and out of context

[avesed]

Description

Keep getting 2 errors when using GPT models. Both login (OpenAI Plus) or using api (from reverse CLI).

1. "Item with id 'rs_xxxxx' not found. Items are not persisted when 'store' is set to false."
2. out of context

Plugins

none

OpenCode version

1.1.12

Steps to reproduce

1. Use gpt-5.2-codex
2. Normal use, after some work
3. error shows

Screenshot and/or share link

Operating System

Ubuntu 24.04 LTS

Terminal

opencode windows

[github-actions]

This issue might be a duplicate of existing issues. Please check:

• GPT-5.1-codex OpenAI item with id not found error #4426: GPT-5.1-codex OpenAI item with id not found error - Same exact error message
• BUG: @ai-sdk/openai 2.0.50 breaks OpenAI OAuth plugin requiring store: false (missing providerOptions in compaction) #3118: @ai-sdk/openai 2.0.50 breaks OpenAI OAuth plugin requiring store: false - Root cause analysis and workaround for the "Items are not persisted when 'store' is set to false" error
• [BUG] Switching between ChatGPT Codex to OpenAI API results in errors #7699: Switching between ChatGPT Codex to OpenAI API results in errors - Similar error when switching between OpenAI models
• Desktop app store data not persisting #7637: Desktop app store data not persisting - Related store persistence issue

Feel free to ignore if none of these address your specific case.

[avesed]

Adding

"store": false

fixes the first issue, but it is still hitting the context limit. I have to manually do /compact to continue.

[rekram1-node]

@avesed can u do opencode export > session.json (this is a command to run in ur terminal and will output the session.json file) can u send me a file of a session that this happened in?

[avesed]

@avesed can u do opencode export > session.json (this is a command to run in ur terminal and will output the session.json file) can u send me a file of a session that this happened in?

I upgraded to 1.1.13, so far (about an hour of use), haven't got any error. But on the other hand, after setting "store": false, it kinda burn though tokens, used about 20M tokens in 40min, in my normal workflow (which don't use this much).

[280601922]

@avesed can u do opencode export > session.json (this is a command to run in ur terminal and will output the session.json file) can u send me a file of a session that this happened in?你能做 opencode export > session.json （这是一个在您的终端中运行的命令，并将输出 session.json 文件）你能发给我一个发生这种情况的会话文件吗？多模型翻译对比

I've encountered this issue as well. The attachment shows the problematic session using the gpt-5.2-codex model. I set store:false, but it didn't resolve the problem.

session.json

[avesed]

@280601922 我的格式是这样的，加limit是因为有时候我会触发issue里的第2个报错，改完之后我确实不会有报错了，但是token用的非常快（尤其如果你用的是反代，直接登录还好一点），建议小心使用

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "openai": {
      "options": {
        "baseURL": "",
        "apiKey": ""
      },
      "models": {
        "gpt-5.2-codex": {
          "options": {
            "store": false,
            "reasoningEffort": "xhigh",
            "textVerbosity": "medium",
            "reasoningSummary": "auto",
            "include": ["reasoning.encrypted_content"],
            "limit": {
            "context": 250000,
            "output": 65000
            }
          }
        }
      }
    }
  }
}

[280601922]

@280601922 我的格式是这样的，加limit是因为有时候我会触发issue里的第2个报错，改完之后我确实不会有报错了，但是token用的非常快（尤其如果你用的是反代，直接登录还好一点），建议小心使用

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "openai": {
      "options": {
        "baseURL": "",
        "apiKey": ""
      },
      "models": {
        "gpt-5.2-codex": {
          "options": {
            "store": false,
            "reasoningEffort": "xhigh",
            "textVerbosity": "medium",
            "reasoningSummary": "auto",
            "include": ["reasoning.encrypted_content"],
            "limit": {
            "context": 250000,
            "output": 65000
            }
          }
        }
      }
    }
  }
}

感谢, 我的opencode.json里的配置跟你的不太一样(有个variants的参数,我理解，是不是options参数是针对用的具体推理级别，额外要指定下? 另外我使用了oh-my-opencode不知道跟这个插件有没有关系

[avesed]

@280601922 这个我不太确定，因为我的opencode没有自己生成opencode.json，我是自己按doc写的~/.config/opencode/opencode.json

[arjun810]

Getting this frequently with codex via zen

[PSU3D0]

Getting this issue probably 50% of sessions - usually after 50-60 turns - with 5.2 Codex

[elijahdev0]

Getting this as well please someone mention if they found a fix