Cadence vs Hatchet
Cadence
cadenceworkflow.io
Hatchet
hatchet.runSide by side
| Figure | Cadence | Hatchet |
|---|---|---|
| Price from | Free | Free |
| Free plan | Yes | Yes |
| Free trial | — | — |
| Open source | — | — |
| Web | Yes | Yes |
| Linux | Yes | Yes |
| API | Yes | Yes |
| Self-hosted | Yes | Yes |
| Plans published | 1 | 2 |
| Spec score · Process Orchestration Software | 8.0 (#9) | 8.0 (#10) |
| Spec score · Workflow Orchestration Software | 8.0 (#15) | 8.0 (#16) |
| Deployment | Deployment options include embedded SQLite, Docker Compose, and Kubernetes with Helm; the page also lists a managed Uber cluster for Uber internal teams. | The orchestration engine is available as Hatchet Cloud or self-hosted, while workers run on the customer's own infrastructure. |
| Founded | 2017 | 2024 |
| Security | TLS is configurable on backing service connections, and SDK connections to the frontend support mutual TLS. | SOC 2 Type II is included on every listed pricing tier. |
| Support | The project points users to Stack Overflow, CNCF Slack, GitHub Issues, and community contributors for support. | The Pay-as-you-go plan lists community support; the comparison lists Discord for both Pay-as-you-go tiers and dedicated Slack for Custom. |
| Who it is for | Cadence targets application developers writing workflows and platform engineers operating Cadence as shared infrastructure. | Hatchet says it is particularly suited to AI agents, massive parallelization, and mission-critical workloads. |
Plans
In their own words
Cadence
Cadence has no write-up on Specifiction yet; its figures are above.
Hatchet
Hatchet is a developer platform for building and deploying AI agents, durable workflows, and background tasks. It persists tasks and agent invocations so teams can debug, retry, and replay execution. Workflow controls include automatic retries, scheduled and cron runs, event and webhook triggers, concurrency settings, priorities, and rate limits. SDKs are documented for Python, TypeScript, Go, and Ruby. Hatchet provides real-time workflow and task monitoring, a built-in OpenTelemetry collector, and searchable application logs. Its orchestration engine runs through Hatchet Cloud or can be self-hosted, while workers run on customer infrastructure. Embedded mode supports local development and CI without an account or Docker; CLI, Docker Compose, and Kubernetes options are also available. The Pay-as-you-go plan starts at free, including 1M runs and 1M events per month; usage beyond those amounts is billed at $30 per million runs and $5 per million events. SOC 2 Type II is included on every listed tier. Hatchet says that tasks are dropped and new task creation returns errors when a free-tier limit is exceeded.