How to prepare an Ultralytics model for the use in Savant

Savant natively uses models in the NVIDIA TensorRT format, optimized for a particular hardware platform. However, users do not need to convert models manually; we encourage them to use the ONNX format, which allows Savant to build TensorRT engines internally.

Once built, the models are cached and loaded quickly unless the cache is moved to a GPU of a different GPU family (e.g., Turing to Ampere) or the batch size changes, which causes Nvinfer, used internally by Savant, to rebuild them. You may also want to rebuild the model for a particular GPU, even within the same family, to ensure it is optimal, because, depending on GPU properties, you can get a better-optimized model for that GPU, especially if you allow TensorRT to use more memory. It’s worth trying if you care about performance maximization. In this manual, we walk through exporting an Ultralytics model to the ONNX format for use in Savant.

Continue reading How to prepare an Ultralytics model for the use in Savant

Savant 0.6.0 is Out: the first release to support Blackwell GPUs

As Blackwell GPUs become increasingly used in computer vision and video analytics, demand for their support in Savant has grown significantly. To maintain support for the widely used line of NVIDIA edge (Jetson Orin) and datacenter devices, we decided to switch Savant from using DeepStream 7.0 to a custom-built DeepStream 7.1 with TensorRT 10.9, enabling support for Blackwell GPUs.

The new Savant release received version 0.6.0. The 0.6.x release line is compatible with the previous long-term support 0.5.x line, so users should be able to migrate to newer Blackwell GPUs without code changes. However, particular models may require tweaking and modifications to work with TensorRT 10.9. Another important update in the 0.6.x line is Python 3.12 (in 0.5.x Savant used Python 3.10). This Python version is known to run faster than Python 3.10, which should improve the performance of CPU-bound workloads.

In 0.6.x, we plan to develop Savant in the same evolutionary manner as in 0.5.x: all new features go to 0.6.x, and 0.5 is frozen for new features.

The next significant Savant upgrade will transition from DeepStream 7.1 to DeepStream 8. The timeline is not defined yet, but mostly depends on the NVIDIA roadmap for supporting the entire Jetson line (Orin, Thor), DGX Spark, and top Blackwell GPUs (B300).

To use Savant 0.6.0 on discrete GPUs, you need drivers 570.133.20 or newer; on Jetson, you need JetPack 6.2.

Release notes and Docker images are available on our GitHub at the link. The documentation is updated to address new system requirements.

Do not forget to join our Discord server, where you can ask questions, promote features, and get quick help.

Choosing the technology for a computer vision product in 2025 (Part 1)

I started to apply computer vision tech in 2004 and have watched this field evolve for 20 years. As of 2025, we have built dozens of hands-on computer vision applications. Over time, they grew into the SAVANT framework, now open source.

In this article, I will cover the current and legacy tech stacks for computer vision applications. I hope this helps beginner developers make their first steps and choose their tech wisely, so they can reach their goals and stay scalable, well-performing, and up to date for at least a few years.

My name is Ivan Kudriavtsev, and I am the founder of the Savant Framework, the tech behind dozens of computer vision applications built on NVIDIA hardware. Now let’s dive into the topic!

Continue reading Choosing the technology for a computer vision product in 2025 (Part 1)

New Release Announcement: Retina RTSP Service Now Available!

We’re excited to announce the official release of Retina RTSP Service, a new addition to the Savant platform!

Retina RTSP Service enables seamless integration of real-time video streaming into your Savant-powered AI video pipelines. With Retina RTSP, you can easily ingest RTSP streams, connect them to your Savant services, and build complex computer vision solutions with minimal effort — all while maintaining low latency and high performance.

Continue reading New Release Announcement: Retina RTSP Service Now Available!

Savant 0.5.1: DeepStream 7.0 and new features

After eight months of development, we proudly present you with Savant 0.5.1. The new Savant is based on DeepStream 7.0 (Savant 0.4 is based on DeepStream 6.4) and includes new samples (34 ready-to-use real-world samples), enhancements, and bug fixes.

We recommend using Savant 0.5.x for the production environment and new projects. Version 0.4.x becomes legacy and will not get updates except for critical bug fixes.

Continue reading Savant 0.5.1: DeepStream 7.0 and new features

Choosing A Camera For Computer Vision Wise: Resolution, Image Quality, Lens and Software

Cameras play the most important role in computer vision projects. Often, the quality of the picture captured by a camera influences way more than the sophistication of the computer vision models used in the solution. The truth is that without a proper camera, implementing a state-of-the-art solution matching business needs is often impossible. For example, facial and optical character recognition applications require high-quality, expensive cameras to deliver a pixel-perfect image to the system. Often, computer vision engineers and product sponsors/owners do not understand the characteristics and meet the situation when the solution is not even possible in the wild.

Continue reading Choosing A Camera For Computer Vision Wise: Resolution, Image Quality, Lens and Software

Savant 0.4.8 is Out: A Bugfix Release

We released a minor update to Savant 0.4.x, including a bug fix found by a user. The bug concerns JSON/image/video sink adapters. The partial fix was implemented in 0.4.7.

The new release is available by the link.

Full Changelog: v0.4.7…v0.4.8

Platform: DeepStream 6.4
JetPack (Jetson): 6.0

Docs: https://docs.savant-ai.io/v0.4.8/

dGPU Images:

docker pull ghcr.io/insight-platform/savant-deepstream:0.4.8-6.4
docker pull ghcr.io/insight-platform/savant-adapters-deepstream:0.4.8-6.4
docker pull ghcr.io/insight-platform/savant-adapters-gstreamer:0.4.8
docker pull ghcr.io/insight-platform/savant-adapters-py:0.4.8
docker pull ghcr.io/insight-platform/savant-deepstream-extra:0.4.8-6.4

Jetson Images:

docker pull ghcr.io/insight-platform/savant-deepstream-l4t:0.4.8-6.4
docker pull ghcr.io/insight-platform/savant-adapters-deepstream-l4t:0.4.8-6.4
docker pull ghcr.io/insight-platform/savant-adapters-gstreamer-l4t:0.4.8
docker pull ghcr.io/insight-platform/savant-adapters-py-l4t:0.4.8
docker pull ghcr.io/insight-platform/savant-deepstream-l4t-extra:0.4.8-6.4