GPT-5 max_completion_tokens

[codereptile]

Before submitting your bug report

• I've tried using the "Ask AI" feature on the Continue docs site to see if the docs have an answer
• I believe this is a bug. I'll try to join the Continue Discord for questions
• I'm not able to find an open issue that reports the same bug
• I've seen the troubleshooting guide on the Continue Docs

Relevant environment info

Description

When using gpt-5, I get:

400 Unsupported parameter: 'max_tokens' is not supported with this model. Use 'max_completion_tokens' instead.

The reason is that only the o-series models apply the patch, gpt-5 doesn't trigger it:

continue/core/llm/llms/OpenAI.ts

Line 143 in ab4bb9c

// a) use max_completion_tokens instead of max_tokens

To reproduce

No response

Log output

[codereptile]

I'm sorry for not filling out all the fields, but I'm extremely sure it's not necessary

[bdougie]

I'm sorry for not filling out all the fields, but I'm extremely sure it's not necessary

Moving forward we are trying to get models support into Discussions for easier discoverability.

[TyDunn]

@codereptile This fix for this is on the way. @RomneyDa will be cutting releases shortly

[TyDunn]

PR: #7054

[RomneyDa]

https://github.com/continuedev/continue/releases/tag/v1.0.21-vscode
This should be available (main release, not prerelease) in the next 20 minutes or so

[codereptile]

Hey! Please don't perceive this as being annoying, but I'm curious what's the ETA to the JB plugin?

[bdougie]

Currently in review #7057

[kberezin-nshl]

Thanks for that! But it is marked as "Pre-release" here on GitHub and it is not available on JB marketplace. Do you have an ETA for that?

[codereptile]

@kberezin-nshl Not optimal, but a solution is to manually download here: https://plugins.jetbrains.com/plugin/22707-continue/versions/eap (make sure you have the EAP version selected, and no, you don't need an EAP IDE itself)

[gmarmstrong]

This issue seems to have returned with the most recent release. I'm getting it on both v1.1.77-vscode and v1.0.22-vscode.

[L03TJ3]

Same here!

[Patrick-Erichsen]

Looking into it now!

[TyDunn]

@Patrick-Erichsen Did you get this fixed?

[Patrick-Erichsen]

Apologies for the late follow up, yes! @L03TJ3 let me know if you're still having issues.

[L03TJ3]

@Patrick-Erichsen No, its working now thanks!
that I still end up using gpt-4o is at least no issue related to continue 🤓