private · Aider
A private backend for Aider
Aider can use any OpenAI-compatible backend. Hushwire gives it the best open coding models through one endpoint that stores no prompts or code.
What changes, what doesn’t
- The agent, your workflow and your tools stay the same.
- The model changes: open-weight coders instead of big-lab models. They are strong on everyday agent work; expect differences on the hardest tasks.
- Where your code goes changes: to Hushwire, then to the open model’s host, with nothing stored on our side.
How your code flows
Config
Shell
# Powered by Hushwire private inference (zero retention): hushwire.vibiz.app
export OPENAI_API_BASE="https://hushwire.vibiz.app/v1"
export OPENAI_API_KEY="hw_your_key_here"
aider --model openai/hush-coder-large --no-show-model-warningsNeed to show a client? The security summary is a one-page answer for questionnaires.
Models
| Alias | Model | Context | Good for |
|---|---|---|---|
| hush-coder | Qwen3 Coder 30B A3B (Qwen) | 262k | Fast, cheap default for edits, refactors and tool loops. |
| hush-coder-large | Qwen3 Coder Next (Qwen) | 262k | The largest Qwen coder. Best for multi-file agent work. |
| hush-kimi | Kimi K2.5 (Moonshot AI) | 262k | Strong agentic coder with long tool chains. |
| hush-glm | GLM 4.7 (Z.ai) | 200k | Reliable tool calling, good at planning and front-end work. |
| hush-deepseek | DeepSeek V4 Flash (DeepSeek) | 128k | Very cheap general coder, good for quick questions. |
Try it before you configure anything
The playground runs the same open coding models. No sign-up, 20 messages a day, and nothing you type is stored.