Integrations
Cline
Cline in VS Code with AI Tokens: the OpenAI Compatible provider, the model to pick and the context window to set.
Last updated: 2026-10-11
Settings
Open Cline's settings with the gear icon and fill in:
| Field | Value |
|---|---|
| API Provider | OpenAI Compatible |
| Base URL | https://api.aitokens.ch/v1 |
| API Key | your key, sk-at-… |
| Model ID | qwen3.8-27b |
Then, under Model Configuration:
| Setting | qwen3.8-27b |
qwen3-coder-30b |
|---|---|---|
| Context Window | 262144 | 32768 |
| Max Output Tokens | 8192 | 8192 |
| Image Support | off | off |
| Computer Use | off | off |
The input and output prices are optional: Cline uses them to show an estimate. What each request cost is in the dashboard.
Which model
qwen3.8-27b. Cline sends its own instructions and tool descriptions with
every request, and each file it reads stays in the conversation. The window of
262,144 tokens leaves room for both. qwen3-coder-30b answers faster and is
enough for small tasks that fit in 32,768 tokens. The other models with tools
are in Which model.
Limits
- 8,192 output tokens per answer, and on
qwen3.8-27bthe reasoning counts in the same 8,192. An answer that writes a long file in one go can be cut: ask Cline to write it in parts. - No reasoning to show. Cline has nothing to display in its thinking
panel: the API does not return the reasoning of
qwen3.8-27b. - Images only on a model that reads them, such as
qwen3.6-35b: switch Image Support on for that model alone.