beta9 README

The project's README from github.com/beam-cloud/beta9. As one markdown file: README.md.

Logo Logo

Run AI Workloads at Scale

Colab ⭐ Star the Repo Documentation Join Slack Twitter AGPL

Beam is a fast, open-source runtime for serverless AI workloads. It gives you a Pythonic interface to deploy and scale AI applications with zero infrastructure overhead.

Watch the demo

✨ Features

📦 Installation

pip install beam-client

⚡️ Quickstart

  1. Create an account here
  2. Follow our Getting Started Guide

Creating a sandbox

Spin up isolated containers to run LLM-generated code:

from beam import Image, Sandbox


sandbox = Sandbox(image=Image()).create()
response = sandbox.process.run_code("print('I am running remotely')")

print(response.result)

Deploy a serverless inference endpoint

Create an autoscaling endpoint for your custom model:

from beam import Image, endpoint
from beam import QueueDepthAutoscaler

@endpoint(
    image=Image(python_version="python3.11"),
    gpu="A10G",
    cpu=2,
    memory="16Gi",
    autoscaler=QueueDepthAutoscaler(max_containers=5, tasks_per_container=30)
)
def handler():
    return {"label": "cat", "confidence": 0.97}

Run background tasks

Schedule resilient background tasks (or replace your Celery queue) by adding a simple decorator:

from beam import Image, TaskPolicy, schema, task_queue


class Input(schema.Schema):
    image_url = schema.String()


@task_queue(
    name="image-processor",
    image=Image(python_version="python3.11"),
    cpu=1,
    memory=1024,
    inputs=Input,
    task_policy=TaskPolicy(max_retries=3),
)
def my_background_task(input: Input, *, context):
    image_url = input.image_url
    print(f"Processing image: {image_url}")
    return {"image_url": image_url}


if __name__ == "__main__":
    # Invoke a background task from your app (without deploying it)
    my_background_task.put(image_url="https://example.com/image.jpg")

    # You can also deploy this behind a versioned endpoint with:
    # beam deploy app.py:my_background_task --name image-processor

Self-Hosting vs Cloud

Beta9 is the open-source engine powering Beam, our fully-managed cloud platform. You can self-host Beta9 for free or choose managed cloud hosting through Beam.

👋 Contributing

We welcome contributions big or small. These are the most helpful things for us:

❤️ Thanks to Our Contributors

beta9 (beam-cloud/beta9)

Searching Database…

1.8K ·GitHub
Beta9
这是做什么的？一个面向无服务器 AI 工作负载的开源运行时——它把 Python 代码跑在可自动扩缩容的 GPU 容器里，空闲时可以缩容到零。它也是 Beam（一个全托管的云服务）背后的引擎。
这是云服务，还是需要我自己部署运行？两者都可以。Beta9 是开源引擎，你可以免费自行部署；Beam 则是构建在其上的全托管云服务，如果你不想自己维护基础设施，用它就行。
最快让它跑起来的方法是什么？最快的方式是用托管云：注册账号，然后跟着 Getting Started 指南走。想自己运行引擎的话，可以按照代码仓库中 Installation 部分的说明自行部署 Beta9。
用它要写 Python 吗，还是需要懂 Go？你只需要写 Python：加上装饰器的 Python 函数会被部署成容器并自动扩缩容。Go 只是引擎本身用的开发语言，和你无关。
自己部署真的免费吗——有什么限制？自行部署是免费的，但许可证是 AGPL-3.0：如果你修改了 Beta9 并把它作为服务对外提供，就必须开源你的代码。通过 Beam 使用托管服务就不用操心这些。
我的工作负载可以用哪些 GPU？在 Beam 的托管云上可以用 4090、H100 等显卡，也可以接入你自己的 GPU。
我到底能用它做什么？主要有三类工作负载：为自定义模型搭建可自动扩缩容的推理接口，运行大模型生成代码的隔离沙箱，以及定时执行的后台任务——README 把它定位为 Celery 队列的替代品。
容器启动有多快？工作负载最多能扩展到多大规模？得益于自定义的容器运行时、调度器和内置缓存，容器可以在一秒内启动。工作负载可以并行扩展到数百个容器，空闲时自动缩容到零。
日常开发有哪些内置功能？热重载、Webhook 和定时任务都是一等公民功能，还可以挂载分布式存储卷，实现跨容器的持久化或共享状态。
What are similar projects?faas, petals, Qwen3.8-Flash-Next-Dual-DGX-Sparks, GLM-5.3-Flash-EXL3-2x-DGX-Sparks-TensorFold, ColossalAI, Model-Optimizer, vllm-omni, airllm
Which topics is it in?faas, developer-productivity, distributed-computing, paas, serverless, agplv3, llm-inference, gpu-acceleration, sandbox, fine-tuning, large-language-models, distributed-systems
Creator
beam-cloud
Size
35.5 MB
Stars
1.8K
Updated
CommitOct 18Releasegateway-0.1.821Oct 16
Age
2y
Commits · 90 Days