| FazBrowse GitHub Viewer | Trending | | Home |
| Tools: [Original HTTPS Page] |
Welcome to the SpacemiT Open Source Upstream Summary page!
This page summarizes the upstream progress of SpacemiT K1/K3 support across open-source projects, including system software, toolchains, AI software, and SpacemiT-maintained independent open-source projects.
It provides:
| Software | Updated | K1 Progress | K3 Progress | Details |
|---|---|---|---|---|
| Linux kernel | 2026-09-09 | Mainline support | Mainline support | Linux kernel Module Status |
| OpenSBI | 2026-07-07 | Completed: major planned support | Planning | OpenSBI Module Status |
| U-Boot | 2026-09-07 | WIP: K1 upstream support | Planning | U-Boot Module Status |
| llama.cpp | 2026-06-05 | WIP | WIP | llama.cpp Module Status |
| FlagGems | 2026-06-05 | WIP | WIP | FlagGems Module Status |
| LLVM | 2026-07-01 | Mainline support | Mainline support | LLVM Upstream Status |
| GCC | 2026-09-01 | Mainline support | Mainline support | GCC Upstream Status |
| Binutils | 2026-09-01 | Mainline support | Mainline support | Binutils Upstream Status |
| Project | Type | Platform | Key | Status | Details |
|---|---|---|---|---|---|
| spine-triton | AI Infra | K1/K3 | Triton CPU backend, MLIR pipeline, RVV/IME/AME | WIP | spine-triton Module Status |
| onnxruntime | AI Infra | K1/K3 | ONNX inference, SpacemiT EP, Python/C/C++ API | WIP | onnxruntime Module Status |
| Reachy Mini | Robotics | K3 | Desktop robot, vision follow, voice & dance | Demo Ready; Docs WIP | Repository |
| Linksee | Robotics | K3 | Wheeled robot, SLAM, navigation | Demo Ready; Docs WIP | Repository |
| LeRobot App | Robotics | K3 | SO101 arm, ACT/SmolVLA, local inference | Demo Ready; Simulation Ready | Repository |
| Humanoid | Robotics | K3 | Humanoid control, MuJoCo, RL inference | Validation | Repository |
Build a UEFI-bootable Ubuntu RISC-V preinstalled image for the SpacemiT K3 Pico-ITX board and flash it through fastboot.
Following the RISC-V IME extension standard, and reusing Vector register resources, these instructions can bring more than a tenfold performance improvement to AI applications at a very small hardw…
| Back | FazBrowse Home | New Git URL |