Aider
AI pair-programming CLI driven by LiteLLM.
Aider is LiteLLM-backed, so any OpenAI-compatible base URL works. Set two env vars and pass `openai/<freetheai-alias>` as the model.
Chat completions Streaming Tool calling
Connect Aider to FreeTheAi.
3 steps
- 1
Set environment variables
Aider reads OPENAI_API_BASE and OPENAI_API_KEY from your shell.
bash export OPENAI_API_BASE=https://api.freetheai.xyz/v1 export OPENAI_API_KEY=fta_PASTE_YOUR_FREETHEAI_KEY - 2
Run aider with the openai/ prefix
LiteLLM uses the `openai/` prefix to route the request through its OpenAI driver. Append your FreeTheAi alias.
bash aider --model openai/glm/glm-5.1 - 3
Optional: persist in .aider.conf.yml
Drop a config in your project root so you do not have to repeat the flags.
yaml openai-api-base: https://api.freetheai.xyz/v1 openai-api-key: fta_PASTE_YOUR_FREETHEAI_KEY model: openai/glm/glm-5.1
Models that pair well with Aider.
olm/deepseek-v4-proglm/glm-5.1olm/kimi-k2.7-code
See the full live catalog at /models. Aliases are stable; pricing on the free tier stays $0.
Common gotchas.
- Always include `/v1` in the base URL.
- LiteLLM may print 'unknown model' warnings. Use `--no-show-model-warnings` or supply context-window via `--model-metadata-file` if it gets noisy.