---
id: smg-grpc-proto
version: "0.4.14"
license: Apache-2.0
license_treatment: permissive
maintenance: active
---

# smg-grpc-proto — SMG gRPC proto definitions for vLLM, TRT-LLM, MLX, TokenSpeed, and SGLang

License: permissive · Maintenance: active · Downloads: 1.4M/mo

## What it is and what it does

smg-grpc-proto is a package of pre-compiled Python gRPC service stubs for the Shepherd Model Gateway (SMG), an orchestration layer for large language model inference. It bundles protocol buffer definitions for four LLM inference backends: SGLang scheduler and encoder services, vLLM engine service, and TensorRT-LLM service. Rather than requiring users to compile proto files themselves, the package ships ready-to-use Python stubs generated at build time.

The package is lightweight—a pure Python wheel with only grpcio and protobuf as runtime dependencies—and targets Python 3.10 and later. It is primarily useful for developers building clients or integrations that communicate with SMG-orchestrated inference services, or for projects like vLLM that need to speak to multiple backend engines via gRPC.

Use it for:

- Building a client application that communicates with SMG-orchestrated vLLM or TensorRT-LLM inference engines.
- Integrating multiple LLM backends (SGLang, vLLM, TRT-LLM) into a single inference orchestration system.
- Testing or developing SMG itself by importing the proto stubs without manually compiling proto files.
- Deploying inference services that use gRPC for inter-service communication in a distributed LLM system.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Provides pre-compiled Python gRPC stubs for protocol buffers used by SMG (Shepherd Model Gateway) to communicate with LLM inference services including SGLang, vLLM, and TensorRT-LLM.

Yes, if you are building or deploying an SMG-based inference system or integrating with vLLM, TensorRT-LLM, or SGLang via gRPC. The package is actively maintained, has no known vulnerabilities, low install friction, and a permissive license. Install only if you actually need these specific proto stubs; it is a narrow-purpose dependency.

## Install

pip install smg-grpc-proto
uv add smg-grpc-proto
poetry add smg-grpc-proto

## Installing smg-grpc-proto

Before you install: Low friction: pure Python wheel with only two runtime dependencies (grpcio and protobuf). Repository is active with recent commits and no archived status.

License in practice: Apache-2.0 permissive license allows commercial and private use with minimal restrictions.

Quickstart:

pip install smg-grpc-proto

from smg_grpc_proto import sglang_scheduler_pb2, sglang_scheduler_pb2_grpc
from smg_grpc_proto import vllm_engine_pb2, vllm_engine_pb2_grpc

Requires Python 3.10 or later; grpcio>=1.81.1 and protobuf>=5.26.0 must be installed.

Verify before relying:

- Whether the pre-compiled stubs are regenerated with each release or lag behind proto source changes
- Compatibility guarantees between this package version and specific vLLM or TensorRT-LLM versions

## Package facts

- License: Apache-2.0 (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 1.4M/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

grpc protocol buffer stubs, llm inference service definitions, vllm trtllm sglang grpc, shepherd model gateway proto, grpc service client stubs, llm engine communication protocol, distributed inference grpc definitions, grpc-stubs, llm-inference, protocol-buffers

[View on SkillFed](https://skillfed.io/packages/smg-grpc-proto) · [View on PyPI](https://pypi.org/project/smg-grpc-proto/)