# Aider v0.58.0: Architect/Editor model pairing

> Introduces splitting a task across two models — a strong reasoning model as Architect and a cheaper, faster model as Editor — with o1-preview and gpt-4o given as the example pairing. Also adds `--o1-preview`/`--o1-mini` shortcuts, Gemini 002 and better Qwen 2.5 support, and renames `/clipboard` to `/paste`.

| | |
|---|---|
| **Tool** | Aider |
| **Version** | v0.58.0 |
| **Kind** | release |
| **Published** | 2024-09-29 |
| **Observed** | 2026-08-11 |
| **Significance** | 4/5 |
| **Breaking** | no |
| **Categories** | feature, capability, model-support, api-change |

> **Note:** this classification is below our confidence threshold and is pending human review.

## What changed


- Architect/Editor model pairing: a strong reasoning model (example: o1-preview) plans, a cheaper faster model (example: gpt-4o) applies the edits
- New `--o1-preview` and `--o1-mini` shortcuts
- Support for the new Gemini 002 models; better support for Qwen 2.5 models
- `/clipboard` renamed to `/paste`; new `/copy` copies the last LLM response to the clipboard
- Many confirmation prompts can be silenced for the session with a "(D)on't ask again" answer
- `/read-only` autocomplete covers the entire filesystem
- New `--voice-format` sends voice audio as wav/mp3/webm
- `ModelSettings` accepts an `extra_params` dict passed to `litellm.completion()`
- Follows HTTP redirects when scraping URLs; cursor shapes in vim mode; new completion-menu color settings
- Numerous bug fixes


## Sources

- [github_release](https://github.com/Aider-AI/aider/releases/tag/v0.58.0) — retrieved 2026-08-11

## Community

_No curated reactions recorded. Facts and community takes are kept in separate layers
and never blended._

---
Canonical: https://changelogs.info/aider/v0-58-0
Entity: https://changelogs.info/aider
Event ID: `evt_2024-09-29_aider_v0-58-0`
Licence: event synthesis © changelogs.info, CC BY 4.0. Linked sources belong to their vendors.