CheapVibeCodeProvider

Bring models from CheapVibeCode into the VS Code language model picker. The extension loads the catalog from the provider API, shows the live token balance, and sends chat requests through the same API with your own key.

Requires VS Code 1.138 or newer.

Setup

1. Install the extension and reload VS Code.
2. Run CheapVibeCodeProvider: Set API Key and paste your CheapVibeCode key. The key is stored in Secret Storage. It is not written to settings, logs, or the catalog view.
3. Run CheapVibeCodeProvider: Open Catalog. The catalog and balance load from https://ru.cheapvibecode.ru/v1 unless you change CheapVibeCodeProvider: Base URL.
4. In chat, pick a model from the CheapVibeCodeProvider group. A model name looks like cheapvibecode: Model Name (x0.5). The multiplier is not rounded.

Until you save a selection for the current base URL, every eligible model is available. After you save a selection, refresh does not add models back. Enable a new model from the catalog.

Catalog

The side panel shows:

• the available token balance and the time it was last checked;
• search, grouping, and sort by multiplier;
• context size, input and output limits, and Tools, Vision, Reason, and Stream badges;
• a local usage log.

The status bar keeps the last successful balance when a later request fails. Click it to open the catalog.

CheapVibeCodeProvider: Refresh Catalog reloads the catalog and balance. CheapVibeCodeProvider: Forget API Key removes the key and clears the loaded catalog. CheapVibeCodeProvider: Reset Usage Stats clears only the local log.

Usage entries store the model, time, duration, status, token counts returned by the API, and an estimated charge of ceil(total tokens × multiplier). They do not store prompts, responses, or the API key. The balance endpoint remains the source of the real balance.

Requests

The extension forwards the chat request to {base URL}/chat/completions and streams the response back. Text, images, tool calls, and tool results are passed through. Reasoning returned by the API is forwarded as extension data because stable VS Code cannot display a native thinking part yet.

Set CheapVibeCodeProvider: Reasoning Effort to send reasoning_effort to models that support it. off sends nothing. The setting applies to every such model; the model picker cannot store it per model yet.

The output limit is the value returned by the API. If the API omits it, the extension uses 128000 and says so in the model tooltip. A model without an input limit or a context window stays in the catalog but is not offered in the picker.

Settings

Commands

• CheapVibeCodeProvider: Set API Key
• CheapVibeCodeProvider: Forget API Key
• CheapVibeCodeProvider: Refresh Catalog
• CheapVibeCodeProvider: Open Catalog
• CheapVibeCodeProvider: Reset Usage Stats

Privacy

Your key and prompts go only to the base URL you configure. The extension does not add another service. The local usage log stays in VS Code global state on this machine.

License

MIT License

Copyright (c) 2026 akromiress

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.