API Provider Requests | Consolidated Discussion

[bturcotte520]

Hi Everyone!

Please leave your feature requests regarding API Providers here, to increase visibility and reduce redundancy.

Thanks for being a part of the Kilo Code Community!

[bturcotte520]

From @kavehsfv:

please add an "ON/OFF" option to the reasoning effort, dropdown or a checkbox/radio button to enable/disable reasoning for models which support it like Deepseek v3.1 or GLM 4.6 on openrouter.

https://openrouter.ai/docs/use-cases/reasoning-tokens#enable-reasoning-with-default-config

[bturcotte520]

From @KainoaNewton:

Add Codex CLI as a provider so people can use their openai account tokens in kilo code. Similar to the Gemini CLI, Claude Code, and Qwen Code providers. This would be super super helpful as I have a openai subscription so I get tokens for codex cli, but I want to use Kilo Code without having to buy another subscription or pay for extra tokens.

[bturcotte520]

From @qiuyueovo:

I hope that the model name can be customized in the idea plug-in.
In this way, I can access the custom model I want.

[bturcotte520]

From @nkwelleepiengomejr :

Description

Add first-class integration for the OpenAI Codex CLI as a provider in Kilo.

Proposal

• MCP path (minimal): expose codex mcp in the Provider/MCP catalog.
• Native path (optional): detect codex on PATH and start a session per workspace.

Scope

• UX: surface Codex-proposed file diffs and shell commands; apply/run only after user approval.

Settings

• Model picker, autonomy level, and a zero-data-retention toggle.

[bturcotte520]

From @LaMarCab76:

Could we add CODEX as an API provider in Kilo Code? I use Kilo Code daily, and CODEX now lets Plus/Pro subscribers to use their account instead of the API pricing.

[bturcotte520]

From @georgij-spiridonov:

It would be great to be able to use Kilo as a provider for the embedding model for indexing. This would allow use a single provider and balance for all our work