Use Inference APIs with Cline (VS Code)
Choose the "OpenAI Compatible" provider in Cline settings and paste the base URL, key and a model id.
Settings
| Setting | Value |
|---|---|
| Base URL | https://api.inferenceapis.com/v1 (also /openai/v1) |
| API key | From API Keys; send as Authorization: Bearer … |
| Chat model ids | openai/gpt-oss-120b, deepseek-ai/DeepSeek-V4-Flash, zai-org/GLM-5.3-Flash, meta-llama/Llama-3.3-70B-Instruct-Turbo — all models |
| Audio model ids | openai/whisper-large-v3 (transcription), hexgrad/Kokoro-82M (speech) |
Which model
openai/gpt-oss-120b. Cline needs reliable tool calling and long system prompts. Prices and measured speed: chat models.
Steps
- Open Cline → settings (gear) → API Provider → OpenAI Compatible.
- Base URL:
https://api.inferenceapis.com/v1 - API Key: your Inference APIs key.
- Model ID:
openai/gpt-oss-120b(tool calling required for Cline; GPT-OSS, DeepSeek V4/V4.1 and GLM 5.3 all support it). - Set Context window to 131072 (GPT-OSS) or 1000000 (DeepSeek / GLM) if Cline asks.
Verify
Ask Cline to "list the files in this folder" — a working tool call confirms the setup.
How this guide was checked
Configuration only These settings follow Cline's documented OpenAI Compatible provider. We verified tool calling on every chat model here, which is what Cline depends on, but did not run the extension itself. If a step does not match what you see, tell us.
Gotchas
- Cline sends long system prompts; budget a few thousand tokens per turn.
- If Cline reports the model does not support tools, double-check the model id is one of the chat models, not a speech model.
If something fails
401 — key missing or wrong · 402 insufficient_balance — add credits on Billing · 404 model_not_found — check the id against the model list (aliases such as gpt-oss-120b work too) · 503 model_unavailable — the model is being enabled. Full details in the API docs.