Support Matrix
Hardware, software, and build compatibility for Dynamo
DocumentationSee also: Release Artifacts for container images, wheels, Helm charts, and crates | Feature Matrix for backend feature support
Latest stable release: v1.1.1 — SGLang 0.5.10.post1 (NIXL 1.0.1) | TensorRT-LLM 1.3.0rc11 (NIXL 0.10.1) | vLLM 0.19.0 (NIXL 0.10.1)
Experimental release: v1.2.0-deepseek-v4-dev.2 (DeepSeek-V4-Flash / V4-Pro on Blackwell, vLLM + SGLang containers only) — vLLM 0.20.0 | SGLang upstream deepseek-v4-blackwell preview | NIXL 0.10.1
On this page: Backend Dependencies | CUDA and Drivers | Hardware | Platform | Cloud | Build Support
Driver requirements differ by backend — see CUDA and Driver Requirements below.
The following table shows the backend framework versions included with each Dynamo release:
For v1.1.0-dev.2, v1.1.0-dev.3, and v1.2.0-deepseek-v4-dev.2, the cells above match container/context.yaml on the corresponding release branch (pins used to build images). Those lines are partial releases: not every backend has a published Dynamo runtime container for that tag. See Pre-Release Artifacts for what actually shipped. The v1.2.0-deepseek-v4-dev.2 SGLang container is built on the upstream lmsysorg/sglang:deepseek-v4-blackwell preview image rather than a tagged SGLang release; TensorRT-LLM is not part of that dev release.
ai-dynamo[trtllm] wheel will fail on Python 3.11.Dynamo container images include CUDA toolkit libraries. The host machine must have a compatible NVIDIA GPU driver installed.
Patch versions (e.g., v0.8.1.post1, v0.7.0.post1) have the same CUDA support as their base version.
Experimental v1.1.0-dev.* images follow the same CUDA matrix as v1.0.2. The v1.2.0-deepseek-v4-dev.2 vLLM container is CUDA 13.0 multi-arch; the SGLang containers split by arch (CUDA 12.9 on amd64, CUDA 13.0 on arm64).
Experimental CUDA 13 images are not published for all versions. Check Release Artifacts for availability.
For detailed artifact versions and NGC links (including container images, Python wheels, Helm charts, and Rust crates), see the Release Artifacts page.
For detailed information on CUDA driver compatibility, forward compatibility, and troubleshooting:
For extended driver compatibility beyond the minimum versions listed above, consider using cuda-compat packages on the host. See Forward Compatibility for details.
Dynamo provides multi-arch container images supporting both AMD64 (x86_64) and ARM64 architectures. See Release Artifacts for available images.
If you are using a GPU, the following GPU models and architectures are supported:
Dynamo is compatible with the following platforms:
Wheels are built using a manylinux_2_28-compatible environment and validated on CentOS Stream 9 and Ubuntu (22.04, 24.04). Compatibility with other Linux distributions is expected but not officially verified.
AL2023 TensorRT-LLM Limitation: There is a known issue with the TensorRT-LLM framework when running the AL2023 container locally with docker run --network host ... due to a bug in mpi4py. To avoid this issue, replace the --network host flag with more precise networking configuration by mapping only the necessary ports (e.g., 4222 for nats, 2379/2380 for etcd, 8000 for frontend).
For version-specific artifact details, installation commands, and release history, see Release Artifacts.
Dynamo currently provides build support in the following ways:
Wheels: We distribute Python wheels of Dynamo and KV Block Manager:
Dynamo Container Images: We distribute multi-arch images (x86 & ARM64 compatible) on NGC:
Helm Charts: NGC hosts the helm charts supporting Kubernetes deployments of Dynamo:
Rust Crates:
Once you’ve confirmed that your platform and architecture are compatible, you can install Dynamo by following the Local Quick Start in the README.