Hey everyone!
This past year brought major progress for Nira, marking a period of focused development, important technical breakthroughs and meaningful growth across the entire platform.
We strengthened the core of the platform, introduced important improvements to rendering quality and performance, expanded support for the key data types used across reality capture, surveying and construction, and grew our team to accelerate development.
We also realized that while we were deeply focused on building the technology, we were not always the best at sharing what was happening behind the scenes. This recap is our first step toward changing that. It also marks the first entry in our newly launched Nira Blog, where we will share regular product updates, technical insights, customer stories and a clearer view into how the platform continues to evolve.
As we look back at the work completed this year, several updates stand out as major steps forward for the platform. Here is a closer look at the improvements that shaped Nira throughout 2025:
High Resolution Point Clouds
As point clouds continue to be the preferred format for professionals working with LiDAR and large scale spatial data, we made it a priority to bring high quality point cloud visualization to Nira this year. The feature was implemented and continuously refined to support larger and more detailed datasets, with ongoing improvements that made it possible to work with increasingly massive scanning projects while maintaining clarity and smooth interaction.
Ultra High Resolution Panoramic Images
Building on our philosophy of supporting unlimited resolution across all major data types, we added ultra high resolution panoramic imagery to Nira this year. The platform now supports panoramas up to 64k by 32k, extending the same approach that powers our unlimited resolution point clouds and large scale texture support.
Nira is the only platform in its category able to load images of this size instantly and render them smoothly in the browser. This is made possible through custom low level rendering and streaming work developed in house and an engineering magic voodoo trick that lets these massive images run at blazing fast speeds.
3D Gaussian Splats
Gaussian Splats became one of the most talked-about technologies this year, and we fully embraced the momentum. Nira now supports large scale splat datasets with effectively unlimited counts, allowing millions of splats to be rendered in real time without performance loss.
We pushed the format far beyond the typical proof-of-concept stage, enabling splats to capture fine detail, smooth surfaces and subtle lighting transitions at massive scale. This update establishes splats as a first class data type in Nira, standing alongside meshes and point clouds with the same level of reliability and visual precision.
Compare Mode
To finish the year off, we introduced one of the most requested features from our users: Compare Mode. This new mode allows users to switch between versions or datasets and evaluate differences without any local processing or manual exporting. All comparisons run server side for consistent performance across devices, regardless of dataset size.
Why Nira Supports Three Primary Data Types
Nira is built around three core formats: textured meshes, point clouds and Gaussian splats. These formats represent the majority of real world 3D workflows. The platform is designed to load each format instantly on any device and to allow users to combine them within the same project. This architecture keeps the platform flexible, scalable and useful for a broad range of industries.
The foundation built this year makes it easier to introduce new viewing modes, performance improvements and cross-format tools. It also supports the next phase of our roadmap, which includes more advanced comparison features, more accurate color and lighting controls and improved multi-dataset workflows.
Key Benchmarks Reached in 2025
Some of the largest datasets processed in Nira this year included more than XX billion points [need exact info] and over xx million splats, both rendered in real time. These benchmarks highlight the scale the platform can handle and serve as a reference point for the performance improvements planned ahead.
Looking Ahead
A more detailed roadmap update and a webinar are planned for early next year. These will cover progress in point cloud rendering quality, new visual and inspection tools, improved version comparison workflows and expanded support for complex multi-format projects. Our goal is to continue increasing clarity, performance and reliability across all major data types.
2025 Recap Video
Below is the full recap video summarizing this year’s progress. It highlights the updates described above and provides a visual overview of how the platform has evolved.
[Embed recap video here]
