Use TokenAir with the AI tools already in your workflow.
TokenAir can lower model spend significantly when a compatible tool moves suitable work to lower-cost models and the result still meets your quality bar. Check compatibility first. Some tools accept a custom endpoint; others do not.
Compatibility at a glance
A direct label means the tool documents the endpoint pattern used by TokenAir. It does not mean every model or agent feature is identical. Test the workflow before moving production work.
OpenCode
Direct setup is available. OpenCode documents a custom provider flow based on the OpenAI-compatible AI SDK package and a configurable baseURL.
Read OpenCode guideHermes Agent
Direct setup is available. Hermes Agent documents an interactive custom endpoint flow and a model.base_url configuration for services that expose /v1/chat/completions.
Read Hermes Agent guideOpenClaw
Direct setup is available. OpenClaw documents custom providers under models.providers with baseUrl, apiKey, api, and models fields.
Read OpenClaw guideWorkBuddy
Direct setup is available through WorkBuddy's visual Custom model settings. Its standard mode appends and validates the /chat/completions path.
Read WorkBuddy guideCursor
A limited setup path exists through Cursor's OpenAI API key and Override OpenAI Base URL settings. Treat it as a chat-model experiment, not a replacement for every Cursor model or feature.
Read Cursor guideAider
Aider's documented configuration maps to TokenAir: OPENAI_API_BASE sets the endpoint, OPENAI_API_KEY stores the credential, and openai/model-name selects the model.
Read Aider guideCline
Cline has the endpoint fields TokenAir needs. Full Agent behavior also depends on the model's function calls, streaming, and computer-use support, so this guide keeps the integration limited.
Read Cline guideRoo Code
Roo Code's Base URL, key, and model fields match TokenAir's endpoint. Roo has no XML tool fallback, which leaves the integration limited until streamed native tool events pass a real task.
Read Roo Code guideContinue
Continue can route chat, edit, and apply roles to TokenAir through provider: openai and apiBase. Agent mode remains limited until a native tool call completes the full round trip.
Read Continue guideOpenHands
OpenHands exposes the Custom Model, Base URL, and API Key fields needed for TokenAir. Agent reliability depends on the selected model and remains unverified through this route.
Read OpenHands guideKilo Code
Kilo Code's custom provider accepts the OpenAI Compatible protocol, Base URL, key, and model list. TokenAir has not yet verified the full file and terminal tool loop, so Agent use remains limited.
Read Kilo Code guideZed Agent
Zed accepts an API URL, key, model ID, context window, and capability settings for a custom provider. Tool behavior through TokenAir remains model-specific and unverified.
Read Zed Agent guideGoose
Goose documents a custom OpenAI host and Chat Completions base path. TokenAir has not yet verified native tool calls and multi-turn tool results for each model.
Read Goose guideHow the cost reduction works
- Connect one compatible client to TokenAir.
- Choose one repeatable, low-risk workflow.
- Test a lower-cost model on the suitable part of that workflow.
- Keep a premium model where quality or risk justifies it.
- Compare cost per accepted result, including retries and fallbacks.
This is why the guides do not promise a fixed saving percentage. A lower token price is useful only when the complete result still passes review.
Check the endpoint before changing a tool.
Verify the TokenAir key, model ID, and Chat Completions response, then follow the tool-specific guide.