Bello

3x more usage on your coding agent, same quality. Have an multi-agent setup in mind? Bello already supports it. Mix models, run agents in parallel, have them ch

What is Bello?

3x more usage on your coding agent, same quality. Have an multi-agent setup in mind? Bello already supports it. Mix models, run agents in parallel, have them check each other’s work, and reduce unnecessary model calls and noisy tool output. Turn on only what you need.

Overview of Bello

What it is

Bello is a tool that builds a cheaper, more reliable coding team from existing models. It places a supervisor, a reviewer, an adversary, and a local log distiller around a coder, each optional and each on its own model. In tests, one configuration used 66.3% less Codex usage at the same quality.

Key features

  • Optional runtime supervisor, completion reviewer, adversary, and log distiller, each independently enabled or disabled and using different models.
  • In tests, the Efficient Budget configuration used 66.3% less of the weekly Codex limit than Raw GPT-5.6 Sol XHigh without lowering average quality, and its mean score was 1.45% higher.
  • Sol Ultra C+A raised the mean score over Raw Codex by 26.4%.
  • Log distiller cut API-equivalent cost by 22% on Astra and by 14.8% on Luna.
  • Goal mode: keeps the run going until the completion reviewer accepts the original objectives, with unlimited completion returns.
  • No Need For Context (NNC): compresses older conversation history into a summary while preserving the original task, instructions, and a recent tail of complete interactions.
  • Smart Execution: runs independent tool calls in parallel and delivers results as they become ready, reducing model calls.

Use cases

  • Reducing model usage on a coding agent while maintaining output quality.
  • Setting up multi-agent systems with different model combinations for coding tasks.
  • Improving code reliability through runtime supervision, completion review, and adversarial testing.
  • Running long tasks with goal-oriented mode or conversation history compression.

Getting started

bash pipx install bello bello doctor Then add model sources (Codex subscription, Claude Code subscription, API providers, or log distiller). For the Codex plugin: bash codex plugin marketplace add AlexeyKulaev/Bello-codex-marketplace --ref main codex plugin add bello@bello-marketplace After installing the plugin, open your coding agent in the project containing task.md and follow the conversation example in the README.

License and pricing

The license is MIT. The project is open source.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/makson179bello.svg)](https://yunodb.com/tools/makson179bello?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/makson179bello?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/makson179bello.svg" alt="Featured on YunoDB" height="20"></a>