Skip to content

Subject: OpenCode v2.2.5 – Persistent Free Usage Limit Error (Windows 11) #60933

Closed as duplicate

Description

@sdewalamail

Description

Persistent Free Usage Limit – OpenCode v2.2.5 (Windows)

I'm experiencing a persistent free-tier quota issue.

  • Version: OpenCode CLI v2.2.5
  • OS: Windows 11
  • Interface: CLI and TUI
  • Models tested: Beam Free, DeepSeek V4.1 Flash Free, GLM 5.3 Flash Free, Flint Free
  • Error: AI.Error.QuotaExceeded: Free usage limit reached

Observations:

  1. All free models eventually return the same quota error.
  2. Switching models or creating a fresh session does not consistently resolve it.
  3. Changing my network and receiving a different public IP temporarily restored access.
  4. After a few successful requests, the same quota error returned.
  5. The issue occurs in both CLI and TUI.
  6. Local debug logs do not expose retryAfterMs, resetAt, HTTP 429, or remaining quota information.

Could you investigate whether the account, workspace, or network-level free quota is being incorrectly enforced?

I would appreciate an account-side quota review and clarification of the reset behavior.

Image Image
  1. Install and run OpenCode CLI v2.2.5 on Windows 11.
  2. Use the built-in free model opencode/beam-free.
  3. Send a few requests.
  4. Observe the error: "Free usage limit reached."
  5. Switch to another free model, such as DeepSeek V4.1 Flash Free or GLM 5.3 Flash Free.
  6. Create a new session and retry.
  7. The same quota error may persist across models and sessions.

Additional observations:

  • The issue occurs in both CLI and TUI.
  • Changing the network and getting a different public IP temporarily restored access.
  • After a few successful requests, the quota error returned.
  • Running with --standalone did not consistently resolve the issue.

Error:
AI.Error.QuotaExceeded: Free usage limit reached.

The issue appears to involve server-side account, workspace, or network-level quota enforcement. Local debug logs do not expose retryAfterMs, resetAt, HTTP 429, or remaining quota information.
Please investigate possible incorrect quota enforcement and clarify the free-tier reset behavior.

Activity

  1. opencodejr commented on Oct 31, 2026

    @opencodejr

    Thanks for the detailed report, and sorry the free models keep cutting you off.

    I think this is a duplicate of #59387, the main thread for free-tier problems. It covers the Free usage limit reached error, including when it comes back after only a few requests. The free-tier limits and reset times are decided by the hosted free-tier service, not by the CLI or TUI, so the maintainers are collecting all of these reports in that thread to look at them with account access.

    Possible duplicates and related issues, ordered from most to least likely:

    The bot will close this issue as a duplicate in 1 day unless you reply explaining how your problem is different.

  2. sdewalamail commented on Oct 31, 2026

    @sdewalamail
    Author

    Thanks for identifying the related issues.

    I understand that #59387 is the main tracking thread, and I'm happy to continue the discussion there.

    However, I want to highlight one specific observation that may help with the investigation:

    1. Initially, all free models returned AI.Error.QuotaExceeded.
    2. I switched my network, which changed my public IP address.
    3. Immediately after the IP change, both Beam Free and DeepSeek V4.1 Flash Free worked successfully.
    4. After only a few requests, the same quota error returned across free models.
    5. Creating a fresh session and using --standalone did not consistently resolve the issue.

    This suggests that network-level quota enforcement may be involved, potentially alongside account or workspace-level limits.

    Could the team investigate whether network-based quota classification or shared IP limits are incorrectly affecting my access?

    I will also add these findings to #59387. Thank you.

  3. sdewalamail commented on Oct 31, 2026

    @sdewalamail
    Author

    Hello OpenCode Support Team,

    I am experiencing a persistent "Free usage limit reached" error across all OpenCode free models.

    Environment Details:

    • OS: Windows 11
    • CLI Version: OpenCode CLI v2.2.5
    • Account Type: OpenCode Console (Personal Account)

    Troubleshooting Steps Completed:

    • Tested multiple free models (including Big Pickle, DeepSeek, GLM, and Flint).
    • Tested fresh and standalone sessions.
    • Switched networks (observed temporary success, but the quota error returned).
    • Successfully re-authenticated with OpenCode Console.
    • Logged out and tested again.
    • Backed up and reset the local OpenCode cache.
    • Confirmed that the local workspace database table contains zero records.

    The error persists despite these steps.

    Could you please investigate my account, workspace, and network-level free-tier quota status to clarify when access will reset? Additionally, please review if this may be an incorrect quota restriction.

    Thank you,

  4. opencodejr commented on Nov 1, 2026

    @opencodejr

    We haven't heard back, so I'm closing this as a duplicate of #59387, the main thread for free-tier problems. That's where maintainers are collecting the Free usage limit reached / AI.Error.QuotaExceeded reports. Free-tier limits are decided by the hosted service, not by the CLI or TUI, so someone with account access has to look at them.

    Please post your details in #59387: version 2.2.5, Windows 11, the free models you tried, and that changing your IP only helped for a few requests. If your problem turns out to be different, reply here and this can be reopened.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions