gRPC
A high-performance open-source Remote Procedure Call (RPC) framework developed by Google, built on HTTP/2 and Protocol Buffers. Donated to the CNCF for stewardship, it is widely used for communication in cloud-native and microservices architectures.
Core Facts
Timeline (last 90 days)
Qdrant v1.19.2 added the ability to bind internal inter-node gRPC sync traffic to a dedicated network interface via PR #10853.
The OTLP protocol choice affects only the transport layer, not data semantics, so switching between gRPC and HTTP carriage does not change metric names or labels.
OTLP/HTTP sends telemetry data via standard HTTP requests, and the default port is usually 4318.
gRPC is a high-performance RPC framework based on HTTP/2 that uses Protocol Buffers to define service interfaces, supports bidirectional streaming and multiplexing, and mandates defining strongly typed method contracts.
ExtMCP is an external gRPC service that agentgateway calls before a tools/call request reaches the upstream server, and calls once more after the response returns.
ML-Agents works as follows: on the Unity side, observation, action, and reward are defined through Agent scripts, and communication with an external Python training process is done via gRPC; on the Python side, reference implementations of algorithms such as PPO and SAC are provided (based on PyTorch), and Stable-Baselines3 or Ray RLlib can also be connected through the mlagents-envs interface.
After logging in, Postman supports more protocol types (such as gRPC and MQTT) and supports managing interface requests through directories; without logging in, it mainly supports the HTTP protocol.
The API world suffers from fragmentation: different styles such as REST, GraphQL, and gRPC, and each vendor's naming conventions, authentication methods, and error code definitions all differ, which is extremely inefficient for AI models.
gRPC uses Protocol Buffers serialization on top of HTTP/2, which can reduce payload size by 30–50% compared to JSON and supports server-side streaming.
Capabilities that need to be dynamically discovered by AI or triggered by natural-language-described intent are suitable for exposure via MCP; fixed machine-to-machine calls between systems and scenarios with strict requirements on latency or throughput should go directly through traditional APIs such as REST or gRPC.
14 more timeline events
All Facts (20)
gRPC is a cross-language RPC framework open-sourced by Google, based on the HTTP/2 protocol and Protocol Buffers serialization.
80%UnverifiedA2A协议的传输层使用JSON-RPC 2.0走HTTPS，长任务用Server-Sent Events做流式更新或Webhook做异步通知，版本0.3还加了gRPC支持
90%Unverifiedrunn支持HTTP/gRPC等协议的接口测试
90%UnverifiedGoogle的API遵循gRPC状态码规范
80%UnverifiedgRPC uses Protocol Buffers by default as its interface definition language and message serialization format.
60%UnverifiedThe OTLP protocol choice affects only the transport layer, not data semantics, so switching between gRPC and HTTP carriage does not change metric names or labels.
50%UnverifiedOTLP/HTTP sends telemetry data via standard HTTP requests, and the default port is usually 4318.
50%UnverifiedgRPC is a high-performance RPC framework based on HTTP/2 that uses Protocol Buffers to define service interfaces, supports bidirectional streaming and multiplexing, and mandates defining strongly typed method contracts.
50%UnverifiedExtMCP is an external gRPC service that agentgateway calls before a tools/call request reaches the upstream server, and calls once more after the response returns.
50%UnverifiedML-Agents works as follows: on the Unity side, observation, action, and reward are defined through Agent scripts, and communication with an external Python training process is done via gRPC; on the Python side, reference implementations of algorithms such as PPO and SAC are provided (based on PyTorch), and Stable-Baselines3 or Ray RLlib can also be connected through the mlagents-envs interface.
50%UnverifiedAfter logging in, Postman supports more protocol types (such as gRPC and MQTT) and supports managing interface requests through directories; without logging in, it mainly supports the HTTP protocol.
50%UnverifiedThe API world suffers from fragmentation: different styles such as REST, GraphQL, and gRPC, and each vendor's naming conventions, authentication methods, and error code definitions all differ, which is extremely inefficient for AI models.
50%UnverifiedgRPC uses Protocol Buffers serialization on top of HTTP/2, which can reduce payload size by 30–50% compared to JSON and supports server-side streaming.
50%UnverifiedCapabilities that need to be dynamically discovered by AI or triggered by natural-language-described intent are suitable for exposure via MCP; fixed machine-to-machine calls between systems and scenarios with strict requirements on latency or throughput should go directly through traditional APIs such as REST or gRPC.
50%UnverifiedStreaming voice typically relies on bidirectional streaming protocols based on WebSocket or gRPC, or HTTP streaming based on chunked transfer encoding.
50%UnverifiedREST remains the dominant API style, GraphQL is increasingly popular in scenarios requiring flexible queries, and gRPC is widely adopted in microservice communication for its high performance and strongly typed contracts.
50%UnverifiedThe gRPC framework uses protobuf as its default interface description language (IDL)
50%UnverifiedThe proto used by vLLM is typically based on Protocol Buffers (protobuf) and is commonly used in conjunction with the gRPC framework.
50%UnverifiedRPC (Remote Procedure Call) is action-centric, with interfaces directly describing callable functions; common implementations include Google's gRPC (based on Protocol Buffers and HTTP/2) and JSON-RPC.
50%UnverifiedThere are two mainstream API interface specifications: REST and RPC, with REST being dominant and RPC holding a smaller market share.
50%