signalGitHub Trending2026-09-25
harry7557558/spirula-studio
Spirula Studio is a cross-vendor 3D Gaussian Splatting trainer that converts photos or video into splats and meshes within a single binary. It runs on NVIDIA, AMD, Intel, and Apple GPUs via Vulkan, trains up to 10 million full-SH Gaussians in 8GB VRAM, and natively supports fisheye and 360-degree cameras without separate COLMAP installation.
- for who
- Developers and researchers in 3D reconstruction and photogrammetry.
- why now
- New editing and rendering features make this cross-vendor Gaussian splat trainer timely for indie makers.
- what changes
- They can train 3DGS models on any major GPU without Python or COLMAP, directly handling 360-degree video.
- to do
- Download the appropriate binary from the Releases page, unzip, and run the GUI; for remote GPUs, use the CLI with 'spirula --help'.
key points
- Cross-vendor Vulkan compute runs on NVIDIA, AMD, Intel, Apple GPUs
- Trains 10M full-SH Gaussians in 8GB VRAM with quantized training
- Native 360 and fisheye support, no COLMAP needed
#3d gaussian splatting#open-source tool#vulkan#cross-platform
score
score 8 out of 10. 0-10: how dense the facts are, multiplied by how much you can do with them after reading. 8+ means the topic's evidence bar is met: benchmarks and availability for a new model, amount and investors for a funding round, revenue figures for a solo-money story. Below 5 an item does not enter the digest. A press release scores 3 or less, a reprint loses 2, anything older than 14 days loses 1, a headline that misleads loses 3.
read the source