This page is a work in progress.
TODO: Edit, Streamline, Format & Style
TODO: Edit, Streamline, Format & Style
定义
网关是 Livepeer 网络中的主要需求聚合层。 它们接受终端客户提供的视频转码和AI推理请求，然后将这些任务分发到配备GPU的协调器网络中。 在早期的 Livepeer 文档中，这个角色被称为广播者。 思维模型From a Cloud Background?
From a Cloud Background?
Running a Gateway is similar to operating an API Gateway or Load Balancer in cloud computing -
it ingests traffic, routes workloads to backend GPU nodes, and manages session flow
without doing the heavy compute itself.
From an Ethereum Background?
From an Ethereum Background?
Running a Gateway is not like running a validator on Ethereum.
Validators secure consensus whereas Gateways route workloads. It’s more akin to a Sequencer on a Layer 2.
Just as a Sequencer ingests user transactions, orders them, and routes them into the rollup execution layer,
a Livepeer Gateway performs the same function for the Livepeer compute network.
Neither? You can still run a gateway!
Neither? You can still run a gateway!
For the rest of us, running a Gateway is like being a film producer.
You take a request, assemble the right specialists, manage constraints,
and ensure the final result is delivered reliably-without doing every task yourself.
什么是网关？
网关是应用程序进入 Livepeer compute 网络的入口。 它们是连接实时AI和视频工作负载与执行GPU计算的协调器的协调层。 它们作为协议和分布式计算网络之间的关键技术层运行。 网关是一个自托管的 Livepeer 节点，它直接与编排者交互，提交任务，处理付款，并暴露直接的协议接口。 Daydream 等托管服务不是网关。 网关负责- 验证请求
- 选择工作者
- 将请求转换为工作者 OpenAPI 调用
- 聚合结果
网关的作用
网关处理运行可扩展、低延迟AI视频网络所需的所有服务级逻辑：-
作业接收
他们从使用 Livepeer API、PyTrickle 或 BYOC 集成的应用程序接收工作负载。 -
功能与模型匹配
网关确定哪些编排器支持所需的 GPU、模型或流水线。 -
路由与调度
他们根据性能、可用性和定价将任务分派给最佳编排器。 -
市场曝光
网关运营商可以发布他们提供的服务，包括支持的模型、管道和定价结构。
Gateway Functions & Services
Learn More About Gateway Functions & Services
为什么网关很重要
随着 Livepeer 过渡到高需求的实时 AI 网络，网关成为关键基础设施。 它们支持：- 适用于 Daydream、ComfyStream 和其他实时 AI 视频工具的低延迟工作流
- 针对计算密集型工作负载的动态 GPU 路由
- 一个去中心化的计算能力市场
- 通过 BYOC 管道模型实现灵活集成
摘要
网关是 Livepeer 生态系统的协调和路由层。它们暴露功能，定价服务，接受工作负载，并将它们分发给编排器进行GPU执行。这种设计实现了可扩展、低延迟、面向AI的去中心化计算市场。 这种架构使 Livepeer 能够扩展为实时AI视频基础设施的全球提供商。Marketplace Content
Marketplace Content
Key Marketplace Features
1. Capability Discovery
Gateways and orchestrators list:- AI model support
- Versioning and model weights
- Pipeline compatibility
- GPU type and compute class
2. Dynamic Pricing
Pricing can vary by:- GPU class
- Model complexity
- Latency SLA
- Throughput requirements
- Region
3. Performance Competition
Orchestrators compete on:- Speed
- Reliability
- GPU quality
- Cost efficiency
- Routing quality
- Supported features
- Latency
- Developer ecosystem fit
4. BYOC Integration
Any container-based pipeline can be brought into the marketplace:- Run custom AI models
- Run ML workflows
- Execute arbitrary compute
- Support enterprise workloads
Protocol Overview
Understand the Full Livepeer Network Design
Marketplace Benefits
- Developer choice - choose the best model, price, and performance
- Economic incentives - better nodes earn more work
- Scalability - network supply grows independently of demand
- Innovation unlock - new models and pipelines can be added instantly
- Decentralization - no single operator controls the workload flow
Summary
The Marketplace turns Livepeer into a competitive, discoverable, real-time AI compute layer.- Gateways expose services
- Orchestrators execute them
- Applications choose the best fit
- Developers build on top of it
- Users benefit from low-latency, high-performance AI