Last updated: October 2026 · Engine 0.5.4 (stable, alpha status)
Every modality below is published and downloadable. The current list is generated from the registry at neurobrix.es/catalog, not maintained here.
| Modality | Status |
|---|---|
| Image generation | Shipped |
| Language models | Shipped |
| Code models | Shipped |
| Vision-language | Shipped |
| Multimodal / omni | Shipped |
| Speech to text | Shipped |
| Text to speech | Shipped |
| Audio-native LLMs | Shipped |
| Upscaling | Shipped |
| Video generation | Shipped |
Hardware that executes today: NVIDIA CUDA (with the Triton branch), Apple Silicon through MPS (compiled mode) and CPU. See Hardware Support.
These are intentions, not commitments. No dates are published.
sha256; the CLI does not check it in 0.5.4.For measured speed and its limits, see Benchmarks. For what changed recently, see the Changelog.
Next: Changelog
© 2025–2026 Hocine Benkelaya. NeuroBrix engine released under the Apache License 2.0.
Developed by Hocine Benkelaya · Supported and funded by Neural Network Holding Ltd · Commercial distribution by WizWorks OÜ
Documentation source: GitLab