NVIDIA Dynamo

| Support Matrix | Guides | Architecture and Features | APIs | SDK |

NVIDIA Dynamo is a high-throughput low-latency inference framework designed for serving generative AI and reasoning models in multi-node distributed environments. Dynamo is designed to be inference engine agnostic (supports TRT-LLM, vLLM, SGLang or others) and captures LLM-specific capabilities such as:

• Disaggregated prefill & decode inference – Maximizes GPU throughput and facilitates trade off between throughput and latency.
• Dynamic GPU scheduling – Optimizes performance based on fluctuating demand
• LLM-aware request routing – Eliminates unnecessary KV cache re-computation
• Accelerated data transfer – Reduces inference response time using NIXL.
• KV cache offloading – Leverages multiple memory hierarchies for higher system throughput

Built in Rust for performance and in Python for extensibility, Dynamo is fully open-source and driven by a transparent, OSS (Open Source Software) first development approach.

Installation

The following examples require a few system level packages. Recommended to use Ubuntu 24.04 with a x86_64 CPU. See support_matrix.md

apt-get update
DEBIAN_FRONTEND=noninteractive apt-get install -yq python3-dev python3-pip python3-venv libucx0
python3 -m venv venv
source venv/bin/activate

pip install ai-dynamo[all]

Building the Dynamo Base Image

Although not needed for local development, deploying your Dynamo pipelines to Kubernetes will require you to build and push a Dynamo base image to your container registry. You can use any container registry of your choice, such as:

• Docker Hub (docker.io)
• NVIDIA NGC Container Registry (nvcr.io)
• Any private registry

Here's how to build it:

export CI_REGISTRY_IMAGE=<your-registry>
export CI_COMMIT_SHA=<your-tag>

earthly --push +dynamo-base-docker --CI_REGISTRY_IMAGE=$CI_REGISTRY_IMAGE --CI_COMMIT_SHA=$CI_COMMIT_SHA

After building, you can use this image by setting the DYNAMO_IMAGE environment variable to point to your built image:

export DYNAMO_IMAGE=<your-registry>/dynamo-base-docker:<your-tag>

Running and Interacting with an LLM Locally

To run a model and interact with it locally you can call dynamo run with a hugging face model. dynamo run supports several backends including: mistralrs, sglang, vllm, and tensorrtllm.

Example Command

dynamo run out=vllm deepseek-ai/DeepSeek-R1-Distill-Llama-8B

? User › Hello, how are you?
✔ User · Hello, how are you?
Okay, so I'm trying to figure out how to respond to the user's greeting. They said, "Hello, how are you?" and then followed it with "Hello! I'm just a program, but thanks for asking." Hmm, I need to come up with a suitable reply. ...

LLM Serving

Dynamo provides a simple way to spin up a local set of inference components including:

• OpenAI Compatible Frontend – High performance OpenAI compatible http api server written in Rust.
• Basic and Kv Aware Router – Route and load balance traffic to a set of workers.
• Workers – Set of pre-configured LLM serving engines.

To run a minimal configuration you can use a pre-configured example.

Start Dynamo Distributed Runtime Services

First start the Dynamo Distributed Runtime services:

docker compose -f deploy/docker-compose.yml up -d

Start Dynamo LLM Serving Components

Next serve a minimal configuration with an http server, basic round-robin router, and a single worker.

cd examples/llm
dynamo serve graphs.agg:Frontend -f configs/agg.yaml

Send a Request

curl localhost:8000/v1/chat/completions   -H "Content-Type: application/json"   -d '{
    "model": "deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
    "messages": [
    {
        "role": "user",
        "content": "Hello, how are you?"
    }
    ],
    "stream":false,
    "max_tokens": 300
  }' | jq

Local Development

Container

To develop locally, we recommend working inside of the container

./container/build.sh
./container/run.sh -it --mount-workspace

cargo build --release
mkdir -p /workspace/deploy/dynamo/sdk/src/dynamo/sdk/cli/bin
cp /workspace/target/release/http /workspace/deploy/dynamo/sdk/src/dynamo/sdk/cli/bin
cp /workspace/target/release/llmctl /workspace/deploy/dynamo/sdk/src/dynamo/sdk/cli/bin
cp /workspace/target/release/dynamo-run /workspace/deploy/dynamo/sdk/src/dynamo/sdk/cli/bin

uv pip install -e .

Devcontainer Environment

For a consistent development environment, you can use the provided devcontainer configuration. This requires:

• Docker
• VS Code with the Dev Containers extension

To use the devcontainer:

1. Open the project in VS Code
2. Click on the button in the bottom-left corner
3. Select "Reopen in Container"

This will build and start a container with all the necessary dependencies for Dynamo development.

Conda Environment

Alternately, you can use a conda environment

conda activate <ENV_NAME>

pip install nixl # Or install https://github.com/ai-dynamo/nixl from source

cargo build --release

# To install ai-dynamo-runtime from source
cd lib/bindings/python
pip install .

cd ../../../
pip install .[all]

# To test
docker compose -f deploy/docker-compose.yml up -d
cd examples/llm
dynamo serve graphs.agg:Frontend -f configs/agg.yaml