Use Kimi K2.6 with Cline in Morocco — velqa.dev, pay in dirhams (MAD)
Configure the Cline VS Code extension to use velqa.dev models with local payment in Moroccan dirhams.
Prerequisites
- VS Code with the Cline extension installed
- A velqa.dev API key (see the quickstart)
Configuration via the Cline interface
Configuration is done entirely from the extension's graphical interface — no need to edit settings.json manually.
- Open Cline in the VS Code sidebar (Cline icon)
- Click the Settings icon (gear) at the top right of the panel
- Under API Provider, select OpenAI Compatible
- Fill in the fields:
| Field | Value |
|---|---|
| Base URL | https://api.velqa.dev/v1 |
| API Key | sk-... |
| Model ID | kimi-k2.6 |
| Context Window Size | 131072 |
| Max Output Tokens | 16384 |
- Click Save
Context window and output (important)
Fill in Context Window Size and Max Output Tokens (in the OpenAI Compatible provider's advanced settings). These two values tell Cline when to automatically compact the context. Without them, a long agentic session lets the context grow unbounded and eventually triggers Trop de tokens demandés. Reessaie plus tard. — while also burning through your quota faster.
Values per model:
| Model ID | Context Window Size | Max Output Tokens |
|---|---|---|
kimi-k2.6 | 131072 | 16384 |
minimax-m3 | 131072 | 16384 |
hy3 | 262144 | 16384 |
glm-4.7-flash | 200000 | 16384 |
deepseek-v4-flash | 65536 | 8192 |
Switching models
To switch to another model, go back to the settings and change the Model ID field:
kimi-k2.6— coding agent, long tasksminimax-m3— code generation and refactoringglm-4.7-flash— fast tasks, general chat — the most responsivedeepseek-v4-flash— high-volume, economical
Test
Open the Cline chat and type a simple instruction, for example Explain this file. If Cline replies, the configuration is correct.
On a 401 Unauthorized error, check that the sk-... key is copied correctly (no spaces). On a Model not found error, check the exact model identifier at velqa.dev/models.
Note on native tools
Cline uses OpenAI native tool calling. Kimi K2.6 and MiniMax M3 support this protocol — they are compatible with no extra configuration.
