Ross ROSS = Recommend OSS · open-source software intelligence for agents

rivet-dev/actors

Rivet Actors are the primitive for stateful workloads. Built for AI agents, collaborative apps, and durable execution. observed · 2026-08-28

github.com/rivet-dev/actors · homepage · Rust · Apache-2.0 (permissive) observed · 2026-08-28

Health v2 · maintenance only

94/100

  • Activity 99
  • Release rhythm 95
  • Longevity 83
How is this computed?

round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-02. Adoption (stars, forks) is never an input.

  • gap_med: 1
  • age_days: 1174
  • days_rel: 38
  • days_push: 7
  • n_releases_24m: 46

Full methodology

Adoption not part of the score

6088 stars · 232 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Rivet Actors is an open-source framework (RivetKit) providing long-running, stateful actor processes as the primitive for AI agents, collaborative apps, and durable execution. It bundles in-memory state with persistence, WebSockets for realtime streaming, durable queues, workflows, and scheduling, with scale-to-zero deployment on Rivet Cloud or self-hosted infrastructure.

Use cases

  • build ai agents with persistent memory and streaming responses
  • run one durable process per user or session
  • build realtime collaborative apps like chat and multiplayer
  • run durable multi-step workflows that survive restarts
  • process background jobs with durable queues
  • schedule cron jobs and timers inside stateful backends
  • self-host durable objects infrastructure like cloudflare durable objects
  • scale stateful workloads to zero when idle

When to choose

  • you need stateful, long-lived backend processes with in-memory speed and automatic persistence
  • you are building AI agents that need memory, queues, and realtime streaming in one primitive
  • you want durable-objects-style actors without being locked to Cloudflare
  • you need scale-to-zero economics for bursty, per-user workloads

When to avoid

  • you only need stateless request/response APIs with no shared state
  • you want a simple managed message queue without compute co-location
  • your team cannot adopt an actor-model programming model
  • you need mature, battle-tested tooling over a newer framework

Facets

framework · maturity active

agent-framework websocket message-queue scheduling state-management workflow-automation rpc http-server sdk large-language-models backend microservices web-development developer-tools self-hosted bun rust cloud self-hosted cross-platform actors durable-execution durable-objects stateful-workloads realtime multiplayer serverless edge-computing cloudflare-workers foundationdb agent-orchestration rivetkit ai-agents real-time nodejs docker kubernetes web-server

10 sources

Member repositories

RepositoryRoleHealth v2
rivet-dev/actorsmain94

For agents

markdown · JSON · MCP: product_card(name="rivet-dev/actors")

Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem