Guide · aider
Connect aider to Synap
-
Get a free key
Sign in with Google at synap.linkrra.com — no card. Your key starts with
sk-pool-. -
Install aider
Use Python 3.12 — aider’s audio dependency does not import on 3.13 yet.
uvx --python 3.12 --from aider-chat aider --version -
Point it at Synap
Aider routes through LiteLLM, so the variable is
OPENAI_API_BASE— notOPENAI_BASE_URL— and the model id takes anopenai/prefix.export OPENAI_API_BASE=https://synap.linkrra.com/v1 export OPENAI_API_KEY=sk-pool-... -
Start it on your files
LiteLLM has no entry for our ids, so silence its unknown-model banner.
aider --model openai/qwen3-coder-30b \ --no-show-model-warnings \ fizz.py
Notes, prices and limits
Two environment variables and an openai/ prefix. Recommended model: qwen3-coder-30b —
$0.0675 in and $0.27 out
per 1M tokens, 98,304-token context.
Run live against this API on aider-chat 0.86.2 under Python 3.12: it applied its edit and the resulting code was executed and asserted correct, not just read.
Aider does not start on Python 3.13 — its pydub dependency imports audioop, removed from the standard library in 3.13, and you get ModuleNotFoundError: No module named 'pyaudioop'. Pin 3.12, or install audioop-lts. This is aider’s own packaging, nothing to do with Synap, but it is the first thing you will hit.
LiteLLM has no entry for our model ids, so aider prints an unknown-model banner and cannot show a context limit. Cosmetic. --no-show-model-warnings clears it; a .aider.model.settings.yml entry gives it the real numbers.
A config file instead of the two variables, if you prefer: .aider.conf.yml with model, openai-api-base and show-model-warnings: false.
On the free tier: every request counts input and output against 100,000 free tokens a day, and answers stop at 4,096 output tokens. Coding agents send long prompts, so the free tier covers a short session; credits cover the rest.
Field names from aider docs: OpenAI-compatible APIs, and aider 0.86.2 run against this API, checked 2026-11-03 20:20 UTC. Prices refresh from /v1/models when this page loads.