Dapr Agents
Framework for resilient agents built on Dapr
Dapr Agents is a Python framework from the CNCF Dapr project for running AI agents as durable, resilient workflows. The DurableAgent persists its state through Dapr Workflows and state stores, so agents resume after restarts. In multi-agent systems, agents communicate event driven via pub/sub, and an orchestrator selects the next agent using a Random, RoundRobin or LLM-based strategy. LLMs are connected via Dapr's Conversation API and tools also via MCP; the project is licensed under Apache-2.0.
Good fit if
Focus: Platform and backend teams running agents reliably in distributed systems
- your project is about Long-running agent workflows, Event-driven agent teams or Agents in microservice architectures
- someone on your team writes code
Probably not if
- nobody on your team writes code (1/5 for use without developers)
- your team is just getting started with agents (skill level: Advanced)
- there is little time for training (learning curve: steep)
Key figures
- Usable without developers
- 1/5
- Scalability
- 5/5
- Integrations
- 4/5
- Versioning
- 2/5
- Observability
- 4/5
Editorial rating on a scale from 1 to 5. Methodology
Fact sheet
- Category
- Developer
- Type
- Framework for resilient agents built on Dapr
- Licence
- Open Source (Apache-2.0)
- Pricing
- Open SourceFree under Apache-2.0. Costs arise for LLM providers and your own infrastructure.
- Deployment
- Self-Hosted, Docker, Kubernetes
- Models
- Dapr Conversation API (anbieterübergreifend)
- RAG
- Yes
- Multi-Agent
- Yes
- SSO, roles (RBAC)
- SSO no, RBAC no
- Latest version
- dapr-agents 1.0.9 (PyPI, 2026-10-09)
- GitHub stars
- 752
Key features
- Multi-agent orchestration via pub/sub (Random, RoundRobin, LLM-based)
- DurableAgent with Dapr Workflows
- State in 28 state store providers
- MCP support
- Observability via OpenTelemetry and Prometheus
Typical use cases
- Long-running agent workflows
- Event-driven agent teams
- Agents in microservice architectures
- Resilient automation
Market notes
Changes to Dapr Agents by email
We let you know when the status, pricing or license of Dapr Agents changes, together with the weekly market notes.
Implementation
Get Dapr Agents implemented
Free for you. Partners only pay for requests they accept.
Alternatives
Similar systems
Similar in category, features or use cases.
- DeveloperAgno (Phidata)Agno, formerly Phidata, is a Python framework for agents and multi-agent teams with memory, knowledge bases and guardrails. AgentOS lets you serve agents as an API in your own cloud, and a control plane is used for testing, tracing and monitoring. The framework is licensed under Apache 2.0, while the hosted control plane is paid.View profile
- DeveloperAtomic AgentsAtomic Agents is a lightweight and modular framework for building agentic AI pipelines and applications. It enables the creation of AI applications with single-purpose, reusable, and composable components. The framework emphasizes control and predictability of outcomes.View profile
- DeveloperAutoGenAutoGen is an open-source framework from Microsoft Research for multi-agent systems, known for group chats between agents and code execution. With the release of Microsoft Agent Framework 1.0 in April 2026, AutoGen entered maintenance mode: it only receives bug fixes and security updates, and new development happens in Agent Framework. Microsoft recommends Agent Framework for new projects. Independently of Microsoft, the community continues the fork AG2.View profile
- DeveloperCAMEL-AICAMEL-AI is an open-source framework and research community focused on multi-agent systems. The framework provides role-playing agents, agent societies, tools for synthetic data generation and simulation environments. It grew out of the 2023 CAMEL paper on communication between agents and supports a large number of model providers.View profile