GPT-5.6 models do not work

Support Bug Reports
,
1

Where does the bug appear (feature/product)?

Cursor IDE

Describe the Bug

Error when I try to send a prompt with any GPT-5.6 model:
Request ID: 314222ab-e0e3-4027-b99a-fbd03f8c9bc6
{“error”:“ERROR_PROVIDER_ERROR”,“details”:{“title”:“Provider Error”,“detail”:“Provider returned error: {"error":{"code":"1214","message":"tools[7].type:type is illegal"}}”,“isRetryable”:false,“additionalInfo”:{},“buttons”:,“planChoices”:},“isExpected”:true}
Provider returned error: {“error”:{“code”:“1214”,“message”:“tools[7].type:type is illegal”}}
Provider Error Provider returned error: {“error”:{“code”:“1214”,“message”:“tools[7].type:type is illegal”}}
NonRetriableError: Provider Error Provider returned error: {“error”:{“code”:“1214”,“message”:“tools[7].type:type is illegal”}}
at dVv (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.glass.main.js:9155:6590)
at EAr (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.glass.main.js:9155:5410)
at Lgm (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.glass.main.js:9156:11974)
at Object.classify (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.glass.main.js:9156:16506)
at mVv (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.glass.main.js:9156:5856)
at async t4a.runWithSharedTurnRunner (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.glass.main.js:9156:15858)
at async Nq.runAgentLoop (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.glass.main.js:20120:3005)
at async uhl.streamFromAgentBackend (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.glass.main.js:20180:14344)
at async uhl.getAgentStreamResponse (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.glass.main.js:20180:23914)
at async GBe.submitChatMaybeAbortCurrent (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.glass.main.js:9326:2476)
at async twm.submitMessage (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.glass.main.js:9213:33107)
at async $7w.submitMessage (vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.glass.main.js:25270:21619)
at async vscode-file://vscode-app/Applications/Cursor.app/Contents/Resources/app/out/vs/workbench/workbench.glass.main.js:22773:80683

Steps to Reproduce

Select any GPT-5.6 model and attempt to send a message.

Expected Behavior

AI should reply.

Operating System

MacOS

Version Information

Version: 3.11.13 (Universal)
VS Code Extension API: 1.125.0
Commit: 3f21b08f0b436a07be29fbfe00b304fa15553350
Date: 2026-07-10T01:45:28.254Z
Layout: Agent Window
Build Type: Stable
Release Track: Nightly
Electron: 40.10.3
Chromium: 144.0.7559.236
Node.js: 24.15.0
V8: 14.4.258.32-electron.0
xterm.js: 6.1.0-beta.256
OS: Darwin arm64 25.5.0

For AI issues: which model did you use?

GPT-5.6 Sol on Medium thinking

For AI issues: add Request ID with privacy disabled

Request ID: 314222ab-e0e3-4027-b99a-fbd03f8c9bc6

Does this stop you from using Cursor

No - Cursor works, but with this issue

5

Hey, I checked the logs for your Request ID. This isn’t a Cursor outage, it’s caused by your BYOK setup.

In Cursor Settings > Models, you have a custom OpenAI Base URL enabled and it points to api.z.ai which is the Zhipu GLM gateway, not OpenAI. Because of that, any GPT model you pick including GPT-5.6 Sol gets sent to z.ai, and it rejects Cursor’s tool schema with this error:

{"error":{"code":"1214","message":"tools[7].type:type is illegal"}}

That’s why non-GPT models work for you like Grok/Composer, Claude Haiku, Gemini, but GPT-5.6 fails. The override only affects OpenAI and GPT models.

Fix: turn off the custom OpenAI Base URL override or disable that BYOK key in Cursor Settings > Models. Then GPT-5.6 will go through Cursor infrastructure and should work normally.

Let me know if you still see issues after that.

6

Me and my team are experiencing the same apparent outage as well. We are both on Team Premium seats and do not have any BYOK configured.

Some examples of failing requests:

Request ID: c5f6c340-2d43-47b6-9710-1d12399829fb
{"error":"ERROR_RESOURCE_EXHAUSTED","details":{"title":"High Load","detail":"We're experiencing high demand for GPT-5.6 Sol right now. Please switch to Auto, another model, or try again in a few moments.","isRetryable":true,"additionalInfo":{},"buttons":[],"planChoices":[]},"isExpected":true}
[resource_exhausted] Error
RetriableError: [resource_exhausted] Error
    at k_f (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9155:6669)
    at pQi (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9155:5404)
    at wfp (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9156:11972)
    at Object.classify (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9156:16501)
    at T_f (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9156:5856)
    at async Dzo.runWithSharedTurnRunner (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9156:15853)
    at async $z.resume (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:20557:7462)
    at async ewa.streamFromAgentBackend (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:20585:14348)
    at async ewa.getAgentStreamResponse (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:20585:23920)
    at async aDe.submitChatMaybeAbortCurrent (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9322:2475)

Second example:

Request ID: 58872996-a929-4fbb-98d7-0f29a1792861
{"error":"ERROR_RESOURCE_EXHAUSTED","details":{"title":"High Load","detail":"We're experiencing high demand for GPT-5.6 Sol right now. Please switch to Auto, another model, or try again in a few moments.","isRetryable":true,"additionalInfo":{},"buttons":[],"planChoices":[]},"isExpected":true}
[resource_exhausted] Error
RetriableError: [resource_exhausted] Error
    at k_f (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9155:6669)
    at pQi (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9155:5404)
    at wfp (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9156:11972)
    at Object.classify (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9156:16501)
    at T_f (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9156:5856)
    at async Dzo.runWithSharedTurnRunner (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9156:15853)
    at async $z.resume (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:20557:7462)
    at async ewa.streamFromAgentBackend (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:20585:14348)
    at async ewa.getAgentStreamResponse (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:20585:23920)
    at async aDe.submitChatMaybeAbortCurrent (vscode-file://vscode-app/usr/share/cursor/resources/app/out/vs/workbench/workbench.desktop.main.js:9322:2475)
7

Seems to be working, thanks! Appreciate it!

8

Mine also seems to be back now as well.