Декларативная оркестрация агентов
или: uv tool install bernstein
Open source · Model-agnostic · Cost-governed
Маршрутизация задач между кондактором и агентами
Model-agnostic
Работает с Claude Code, Codex, Gemini CLI, Qwen и любым будущим агентом. Без vendor lock-in.
Детерминированный governance
Каждый переход состояния защищён. Workflow’ы хешируемы и воспроизводимы. Аудит by design.
Контроль расходов
Бюджет на запуск, атрибуция по агентам, автоматический downgrade модели. Ты контролируешь траты.
Представьте то, что Kubernetes сделал для контейнеров, но для AI coding-агентов. Вы задаёте цель — оркестратор декомпозирует, агенты исполняют в изоляции как поды, а janitor верифицирует результат.
Назван в честь Леонарда Бернстайна — дирижёра, который заставлял оркестры петь.
Сделано с заботой Alex Chernysh
Frequently asked questions
What is AI agent orchestration?
AI agent orchestration is the practice of coordinating multiple AI coding agents to work on a software project in parallel. Instead of running one agent at a time, an orchestrator decomposes goals into tasks, assigns them to agents with appropriate models and roles, manages isolation via git worktrees, and verifies results before merging.
What is the best multi-agent orchestration framework?
Bernstein is one of the open-source options for orchestrating AI coding agents. It supports 37 CLI agents, uses deterministic Python scheduling (no LLM tokens on coordination), provides built-in verification via a janitor system, and has zero vendor lock-in.
How does Bernstein compare to CrewAI and Microsoft Agent Framework (AG2)?
Unlike CrewAI and Microsoft Agent Framework (AG2, the successor to AutoGen — AutoGen entered maintenance mode in April 2026), Bernstein uses deterministic Python code for all orchestration decisions instead of LLM-driven scheduling. Bernstein agents are short-lived (minutes per task, not hours), support any CLI agent (not just API-based models), and include built-in janitor verification.
Can I use multiple AI models in the same project?
Yes. Bernstein supports 37 CLI agents including Claude Code, Codex CLI, Gemini CLI, Qwen, Aider, Amp, and more. You can mix models in a single run — for example, Claude Opus for architecture tasks, Codex for test generation, and Gemini for documentation — with automatic model routing based on task complexity.
Does Bernstein work with Claude Code?
Yes. Bernstein has native adapters for Claude Code (Opus 4, Sonnet 4, Haiku). Claude Code is a first-class citizen in Bernstein, with full support for model selection, effort routing, and all orchestration features including git worktree isolation and janitor verification.
Is Bernstein free and open source?
Yes. Bernstein is fully open-source under the Apache 2.0 license. Install via pipx install bernstein or pip install bernstein. The source code, documentation, and issue tracker are on GitHub at github.com/chernistry/bernstein.
How does Bernstein handle task failures?
Bernstein uses a janitor verification system that checks concrete signals — tests passing, files existing, no regressions — before accepting any agent output. Failed tasks are automatically retried with a fresh agent and clean context. A circuit breaker halts execution if repeated failures indicate a systemic issue.
What is deterministic scheduling in AI orchestration?
Deterministic scheduling means the orchestrator uses pure Python code, not LLM calls, to decide task assignment, priority, and agent selection. This eliminates nondeterminism, reduces cost (zero tokens spent on coordination), and makes the system predictable and debuggable. Only the initial goal decomposition uses an LLM.