
Rivet发布OpenOrange,可在任意服务器上运行Cloudflare Workers
来源标题:Run Cloudflare Workers Anywhere
Rivet推出OpenOrange,让Cloudflare Workers和Durable Objects能运行在用户自选的机器上,突破Worker仅128MB内存等运行时限制,费用更低。它保持与Wrangler一致的开发体验,wrangler.jsonc仍为配置来源,可选择AWS、Google Cloud、Azure或自有服务器部署。
以下正文同步自 Hacker News,版权归原站所有,已转换为易读排版。
AWS
Google Cloud
Azure
Rivet Cloud
Your servers
The same developer experience, without the runtime limits, for a fraction of the bill.
Introducing OpenOrange Watch the walkthrough 7:18
Run it where you want.
Next to your database, in the cloud and region you choose.
AWS Graviton · us-east-1AWS on-demand Graviton (ARM) prices in us-east-1, provisioned 40% above use for autoscaling. Reserved pricing, spot and credits are lower. Cloudflare and AWS list prices checked 2026-10-08.
128 MB is all the memory a Cloudflare Worker gets. OpenOrange runs your Workers and Durable Objects on machines you choose, so the limits are your hardware's.
Workers Paid plan, from Cloudflare's docs. Free plan limits are stricter.
Your wrangler.jsoncBindings, environments, build commands and .dev.vars work as they do with Wrangler. stays the source of truth. Same commands, new name.
Dry run. Nothing was built or sent to Rivet Cloud.
staging.piedpiper.com
/Anton OpenOrange Agents1New
Ask for code. The agent works in its own VM, and everyone in this thread sees it live.
Write fib.py that prints the first 15 Fibonacci numbers, run it, and tell me their sum.
0 1 1 2 3 5 8 13 21 34 55 89 144 233 377 sum: 986
Richard, I wrote fib.py and ran it. It printed the first 15 Fibonacci numbers, starting from 0.
Ask for code. The agent works in its own VM, and everyone in this thread sees it live.
Run middle-out on season-finale.mov and report the Weissman score.
season-finale.mov 4.2 GB → 806 MB weissman score 5.2
Richard, middle-out took season-finale.mov from 4.2 GB to 806 MB.
[{"rest":" dev","scene":"dev","html":"● Ready in 1.4s
Dry run. Nothing was built or sent to Rivet Cloud.
Copy prompt for your agent Copy prompt for your agent Run this project's Cloudflare Worker on Rivet with OpenOrange. First, ask me where it should run: on my machine, on a Rivet cluster, or on my own Rivet Engine. Then read https://github.com/rivet-dev/open-orange/blob/main/README.md and follow its quickstart. Do not change my app code.
1 Step 1: Install
Install the orange CLI globally, then run it in your Worker project. Your code and your Wrangler configuration do not change.
2 Step 2: Run your Worker
Reads your Wrangler configuration. Each Durable Object runs as a Rivet Actor with its own SQLite database.
3 Step 3: Deploy
Builds an image of your Worker and rolls it out to your own cloud (AWS, GCP, Azure) or on-prem servers. Or use Rivet Cloud: a fully managed, dedicated cluster on AWS or Google Cloud, in the region you choose.
orange migrate copies every Durable Object and D1*Cloudflare doesn't provide a way to migrate D1 live, so requests to a D1 database are interrupted while its data copies. This takes a few minutes. database from Cloudflare while your app keeps serving. No maintenance window, no lost writes.
","#F6821F"],"aws":["0 0 304 182","\n\n\n\t\n\t\n\t\t\n\t\t\n\t\n\n",null],"gcp":["0.3 -0.6 34.3 27.9","",null],"azure":["0 0 96 96","\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n",null],"rivet":["0 0 128 128","",null],"users":["0 0 24 24"," ",null],"globe":["0 0 24 24"," ",null],"refresh":["0 0 24 24"," ",null]},"beats":[{"id":"intro","dur":3},{"id":"cutover","dur":3.4},{"id":"lazy","dur":5.6},{"id":"backfill","dur":3.2},{"id":"after","dur":4.6}]}
Supported bindings.
Every Cloudflare binding OpenOrange runs, plus the fetch handler and runtime globals. Each is documented with its status, configuration and known differences from Cloudflare.
For small, spiky Workers, it often is: the free tier and per-request pricing are hard to beat. The meters add up once Durable Objects stay busy. Duration is billed per GB-second while an object is active, and every SQLite row written costs $1 per million. The same GB-hours on Graviton EC2 cost about a fifth as much, and writes cost a fraction of that.
No. OpenOrange implements the Workers API on Node.js and maps each Cloudflare service onto Rivet. It uses Wrangler to read your configuration, and it reuses some MIT-licensed Cloudflare code, such as the Workflows engine. It does not run workerd.
workerd is Cloudflare's runtime, not its platform. It runs your fetch handler, but KV, D1, R2, Queues, Workflows and the Cache API are bindings that call services you have to build, and Durable Objects keep their data on one machine with nothing to place, replicate or scale them. It also keeps limits designed around Cloudflare's network rather than your servers: only the nodejs_compat subset of Node.js, no native npm modules, and alarms capped at 15 minutes. OpenOrange runs your Worker on Node.js with real SQLite, and Rivet provides the services, so your code keeps working on infrastructure you control. Compare OpenOrange and workerd →
No. Rivet is the control plane underneath. You use the orange CLI and your Wrangler configuration.
OpenOrange is Apache 2.0 and free. Self-host Rivet and your cloud bill is the only bill. Rivet Cloud and BYOC pricing is at rivet.dev/pricing.
orange migrate moves Durable Objects while your app serves, with no downtime. D1 databases are interrupted for a few minutes while they copy, because Cloudflare doesn't provide a way to migrate D1 live. R2 is S3-compatible, so rclone or aws s3 sync moves buckets.
It runs wherever you deploy it, in as many regions as you want. Most backends talk to one database; running next to it is faster than running at the edge. For stateless Worker requests, this works fine today. For Durable Objects, an upcoming Rivet update adds multi-region support, including edge regions.
Rivet. OpenOrange is not affiliated with or endorsed by Cloudflare, Inc.
Your Cloudflare project in your datacenter.
正文由 探How 从来源站点 RSS 同步,内容未经改写;遇到排版缺失或需要图片、视频时请以原文为准。