Durable step execution
Wrap existing logic in step.run() so each step gets retries, state, and failure recovery without refactoring the surrounding app.
Execução durável para tarefas em segundo plano, fluxos e agentes de IA
Inngest is a durable execution platform for background jobs, workflows, and AI agents. It moves orchestration concerns such as retries, concurrency, waiting, and observability into the application code while the service handles the execution layer.
The product is designed for teams that want to add reliable step-based logic to existing apps without introducing workers, queues, or a separate orchestration architecture. The source highlights use cases such as long-running workflows, APIs that must survive timeouts, schedules, serverless functions, and event-driven fan-out.
Wrap existing logic in step.run() so each step gets retries, state, and failure recovery without refactoring the surrounding app.
Handle waiting, retries, fan-out, throttling, and scheduling in the platform while the code keeps the orchestration logic close to the function.
Limit the number of steps that run at once, including per-key and cross-function scopes, to protect downstream services and keep tenant workloads fair.
Trace runs with logs, metrics, and replayable history so failures can be inspected at step level and revisited later.
Stream updates from functions in real time, which supports interactive workflows and long-running operations that need progress feedback.
Run background jobs, workflows, and AI agents without adding workers or queues on your side; the platform manages execution infrastructure.
Use Inngest to run multistep backend jobs that should resume after failures instead of restarting from scratch.
Build AI agent flows that can pause for input, wait for events, and continue from the same execution state.
Protect downstream services by capping concurrent steps globally, per tenant, or across multiple functions.
Run workflows that fan out many event-driven tasks and keep each execution traceable and replayable.
Support long-running serverless endpoints or scheduled work that needs reliable retries and visibility.
Inngest runs durable functions by wrapping existing code in steps, so you can add retries, waiting, and orchestration without rebuilding your app around queues or workers.
The pricing page shows a free Hobby plan, a Pro plan that starts at $75 per month, and an Enterprise option with contact-sales pricing.
Yes. The pricing page says Inngest is open-source and can be self-hosted, while the cloud product adds managed infrastructure and observability.
The source highlights durable workflows, background jobs, AI agents, long-running API endpoints, schedules, serverless functions, and event fan-out as common uses.
Concurrency control can be applied with global limits, per-key limits, and shared scopes across functions, which is useful when you need to protect downstream systems or manage multi-tenant load.
Os dados de tráfego são apenas para referência.
hatchet.run
Orquestração de workflows para agentes de IA, tarefas em segundo plano e fluxos duráveis
1code.dev
1Code é um cliente visual de código aberto para agentes de programação que combina Claude Code e Codex, com fluxos para desktop e web, revisão e merge no Git e execuções em segundo plano em sandboxes na nuvem.
browser-use.com
Browser Use automatiza tarefas web para agentes de IA com navegação furtiva, execução hospedada e fluxos escaláveis.
trigger.dev
Plataforma de código aberto para jobs em segundo plano, fluxos de IA e agentes em TypeScript
www.agentconnect.md
AgentConnect permite que as equipes integrem agentes ACP como Claude Code, Codex e Gemini CLI ao Slack, Telegram, Discord e GitHub, para que eles possam assumir tarefas nos mesmos canais em que as equipes já colaboram. É open source e pode ser hospedado por conta própria, com funções, ferramentas, memória e gatilhos configuráveis para fluxos de trabalho com agentes.
venice.ai
Venice é uma plataforma privada de IA para chat, imagens, vídeo, áudio, código, pesquisa e fluxos de trabalho com agentes. Oferece um plano gratuito, planos pagos e uma API compatível com OpenAI com vários modos de privacidade.