@seclink: Sharing https://github.com/XiaomingX/awesome-rk3588…, feel free to star if interested.

X AI KOLs Following Tools

Summary

A curated GitHub repository providing resources for the Rockchip RK3588 SoC, including development boards, communities, system images, kernels, NPU/AI tools, and more.

Sharing https://github.com/XiaomingX/awesome-rk3588…, feel free to star if interested.
Original Article
View Cached Full Text

Cached at: 09/20/26, 01:05 AM

Share https://github.com/XiaomingX/awesome-rk3588…, friends who are interested can give it a star.


XiaomingX/awesome-rk3588

Source: https://github.com/XiaomingX/awesome-rk3588

Awesome RK3588

English | 简体中文

Curated Rockchip RK3588 / RK3588S / RK3588J ecosystem resource navigation: development boards, communities, system images, mainline kernel, NPU/AI, multimedia, UEFI, and toolchains.

RK3588 is Rockchip’s flagship 8K SoC: 4x Cortex-A76 + 4x Cortex-A55 (big.LITTLE), Mali-G610 MP4 GPU, 6 TOPS NPU, supporting 8K@60fps decoding. Unlike the Raspberry Pi with its single forum, it has a decentralized ecosystem of “multiple board communities + one big Armbian circle.” This list centralizes these entry points.

One-line guide: For daily development, check the vendor forums; for system/kernel, check Armbian; for resource indexing, check Awesome RK3588.

Table of Contents

Official Resources

  • Rockchip Official Website (https://www.rock-chips.com/) - Chip specifications, datasheets, official announcements
  • rockchip-linux (GitHub) (https://github.com/rockchip-linux) - Official kernel, u-boot, MPP, buildroot, etc.
  • airockchip (GitHub) (https://github.com/airockchip) - Official NPU/AI-related repository collection
  • RK3588 Brief Datasheet (PDF) (https://www.rock-chips.com/uploads/pdf/2022.8.26/191/RK3588%20Brief%20Datasheet.pdf) - Official specification document
  • Rockchip SDK Resource Download Site (https://console.zbox.filez.com/l/I00fc3) - SDK downloads (access code rknn)

Development Boards & Vendors

VendorRepresentative BoardsForum / Documentation
FireflyITX-3588J, ROC-RK3588S-PCForum (https://forum.t-firefly.com/) · Wiki (https://wiki.t-firefly.com/)
Orange PiOrange Pi 5 / 5 Plus / 5 ProOfficial Site (http://www.orangepi.org/) · Forum (http://www.orangepi.org/orangepibbsen/)
RadxaROCK 5A / 5B / 5TOfficial Site (https://radxa.com/products/rock5) · Documentation (https://docs.radxa.com/rock5)
EMMCube LubanCatLubanCat 4 / 5Documentation (https://doc.embedfire.com/) · Community (https://forums.embedfire.com/c/lubancat/21)
ALIENTEKATK-DLRK3588Openedv (https://www.openedv.com/)
iTOPiTOP-RK3588Official Site (https://www.topeet.com/) · Forum (http://bbs.topeetboard.com/)
ArmSoMSige5 / AIM7Official Site (https://www.armsom.org/)
NeardiLPA3588Official Site (https://www.neardi.com/)
Banana PiBPI-M7Official Site (https://www.banana-pi.org/)
FriendlyElecNanoPC-T6Official Site (https://www.friendlyelec.com/)
Pine64QuartzPro64Official Site (https://www.pine64.org/)

When choosing a board, prioritize its official forum over general communities for quicker access to usable images and patches.

Communities & Forums

Domestic (China)

  • Elecfans · Rockchip Developer Community (https://bbs.elecfans.com/forum.php?mod=forumdisplay&fid=1179) - Resource posts, schematics, SDKs, Android/Linux switching; focused on the original vendor ecosystem and solution providers; highest discussion frequency
  • Firefly Official Forum (https://forum.t-firefly.com/) - Support for Firefly boards, fast responses
  • Radxa Chinese Community (https://forum.radxa.com/) - ROCK 5 series
  • EMMCube LubanCat Community (https://forums.embedfire.com/c/lubancat/21) - Most user-friendly Chinese documentation
  • ALIENTEK Openedv (https://www.openedv.com/) - Tutorials + QQ groups (common RK3588 group numbers: 775783660, 966712463)
  • iTOP Forum (http://bbs.topeetboard.com/) - iTOP-3588 support
  • Zhihu / Bilibili - Good for experience posts, troubleshooting, system installation, running RKNN/RKLLM, but not ideal for asking specific driver questions

International

  • Armbian Forum (https://forum.armbian.com/) - Mainline/near-mainline Linux, images, kernels, NICs, HDMI, NPU compatibility is most concentrated; many ROCK 5 / Orange Pi 5 users
  • Radxa Forum (https://forum.radxa.com/) - Large ROCK 5B/5T user base; concentrated discussions on PCIe, NVMe, 2.5G NIC issues
  • Collabora Blog (https://www.collabora.com/news-and-blog/) - RK3588 upstream progress (HDMI, VPU, ISP, GPU Panthor)
  • LKML (Linux Kernel Mailing List) (https://lore.kernel.org/linux-rockchip/) - linux-rockchip mailing list; track mainline driver progress
  • GitHub Discussions / Issues (https://github.com/) - edk2-rk3588, various board kernel/Debian/Android repositories; very active

Instant Messaging

  • Telegram UEFIonRockchip (https://t.me/UEFIonRockchip) - UEFI / Windows installation / EDK2 tinkering
  • Discord - Windows on ARM / UEFI on Rockchip related servers

Community Quick Reference

What You Want to DoWhere to Go
Buy a board, flash official images, ask about interfacesCorresponding vendor forum
Flash Armbian / mainline kernelArmbian + Radxa
RKNN, cameras, encodingVendor forums + Elecfans
Run LLMs / open-source imagesGitHub Issues + Armbian
Install Windowsedk2-rk3588 + Discord/Telegram
Follow mainline driver progressCollabora blog + LKML

Operating Systems & Images

  • Armbian (https://www.armbian.com/) - Officially supports RK3588, near-mainline kernel
  • Ubuntu Rockchip (Joshua-Riek) (https://github.com/Joshua-Riek/ubuntu-rockchip) - Ubuntu 22.04/24.04 LTS; historically significant, maintenance status has changed considerably
  • DietPi (https://dietpi.com/) - Lightweight Debian derivative
  • Buildroot (rockchip-linux) (https://github.com/rockchip-linux/buildroot) - Minimal embedded system
  • NixOS RK3588 (https://github.com/ryan4yin/nixos-rk3588) - NixOS for Orange Pi 5 / ROCK 5A
  • Arch Linux ARM (https://archlinuxarm.org/) - General-purpose ARM distribution
  • OpenWrt (https://openwrt.org/) - Soft router / gateway
  • Batocera (https://batocera.org/) - Retro gaming system, supports RK3588
  • Android (AOSP) (https://source.android.com/) - Vendor Android 12/13/14 BSPs

Kernel & Mainline Development

  • rockchip-linux/kernel (https://github.com/rockchip-linux/kernel) - Official BSP kernel (includes NPU/VPU drivers)
  • linux-rockchip (https://github.com/armbian/linux-rockchip) - Armbian-maintained near-mainline kernel
  • torvalds/linux (https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/) - Mainline kernel; arch/arm64/boot/dts/rockchip/
  • Collabora: RK3588 mainline status (https://www.collabora.com/news-and-blog/) - Mainline progress on HDMI RX/TX, VPU, ISP, DP, etc.
  • Panthor (Mali Gen10 GPU driver) (https://gitlab.freedesktop.org/mesa/mesa) - Open-source driver for Mali-G610
  • U-Boot (rockchip) (https://github.com/rockchip-linux/u-boot) - Official bootloader

NPU & AI Deployment

  • RKNN-Toolkit2 (https://github.com/airockchip/rknn-toolkit2) - Model conversion and quantization (PC)
  • RKNN-Toolkit-Lite2 (https://github.com/airockchip/rknn-toolkit2/tree/master/rknn-toolkit-lite2) - On-device Python API
  • RKNPU2 (https://github.com/rockchip-linux/rknpu2) - NPU runtime library and C/C++ API
  • RKNN Model Zoo (https://github.com/airockchip/rknn_model_zoo) - Official collection of 30+ model examples (YOLO, SAM, Whisper, CLIP, etc.)
  • RKNN-Toolkit2 Docker (https://github.com/airockchip/rknn-toolkit2/tree/master/docker) - Pre-configured conversion environment
  • RKNPU2 API Reference (PDF) (https://github.com/rockchip-linux/rknpu2/blob/master/doc/Rockchip_RKNPU_User_Guide_RKNN_API_V2.0.0.pdf) - Official API manual
  • View NPU load: cat /sys/kernel/debug/rknpu/load

Large Language Models

  • RKLLM (rknn-llm) (https://github.com/airockchip/rknn-llm) - Official LLM deployment stack; supports Qwen, TinyLlama, Phi, ChatGLM, etc.
  • RKLLM Examples (https://github.com/airockchip/rknn-llm/tree/main/examples) - On-device chatbot examples
  • rkllama (https://github.com/NotPunchnox/rkllama) - RK3588 local inference service compatible with Ollama API

Multimedia & Video Codecs

  • MPP (Media Process Platform) (https://github.com/rockchip-linux/mpp) - Hardware encode/decode base library
  • ffmpeg-rockchip (https://github.com/nyanmisaka/ffmpeg-rockchip) - FFmpeg hardware-accelerated fork
  • gstreamer-rockchip (https://github.com/rockchip-linux/gstreamer-rockchip) - GStreamer plugins
  • librga (https://github.com/airockchip/librga) - 2D graphics/image acceleration
  • rga (rockchip-linux) (https://github.com/rockchip-linux/linux-rga) - RGA kernel driver
  • rkmpi (https://github.com/rockchip-linux/rkmpi) - Video capture/encoding middleware

GPU & Graphics

  • Mesa Panthor (https://gitlab.freedesktop.org/mesa/mesa) - Open-source driver for Mali-G610 (Gen10)
  • Mali Official Driver (libmali) (https://github.com/rockchip-linux/libmali) - Closed-source blob
  • Rockchip Graphics Documentation (https://github.com/rockchip-linux/docs) - DRM / Wayland / X11 explanations
  • Rockchip Vulkan/OpenCL Drivers (https://github.com/rockchip-linux/libmali) - Various version blob archives

UEFI & Windows on ARM

  • edk2-rk3588 (https://github.com/edk2-porting/edk2-rk3588) - RK3588 UEFI firmware (for Windows installation / multi-system boot)
  • Windows on R (https://worproject.com/) - Windows on ARM community; drivers and installation guides
  • Windows on R Docs (https://github.com/worproject/Rockchip-Windows-Drivers) - Windows drivers for various boards
  • UEFIonRockchip (Telegram) (https://t.me/UEFIonRockchip) - Real-time discussion on Windows installation / EDK2

Flashing & Development Tools

  • rkdeveloptool (https://github.com/rockchip-linux/rkdeveloptool) - Linux/macOS command-line flashing
  • rkflashtool (https://github.com/linux-rockchip/rkflashtool) - Open-source flashing tool
  • RKDevTool (https://github.com/rockchip-linux/rkdeveloptool) - Windows GUI flashing tool (Maskrom mode)
  • rkbin (https://github.com/rockchip-linux/rkbin) - Official loader / firmware binaries
  • upgrade_tool (https://github.com/rockchip-linux/rkdeveloptool) - Rockchip official upgrade tool
  • Maskrom / Loader mode: Enter by holding the MASKROM key or shorting the eMMC clock

Projects & Examples

  • rknn_model_zoo (https://github.com/airockchip/rknn_model_zoo) - Official model example collection
  • yolov8_rknn_Cplusplus (https://github.com/cqu20160901/yolov8_rknn_Cplusplus) - Production-grade YOLOv8 C++ inference
  • rknn-multi-threaded (https://github.com/leafqycc/rknn-multi-threaded) - Multi-threaded NPU inference; ~40% performance boost
  • yolov5_Deepsort_rknn (https://github.com/Zhou-sx/yolov5_Deepsort_rknn) - Real-time object tracking
  • rknpu2 rknn_api_demo (https://github.com/rockchip-linux/rknpu2/tree/master/rknn_api_demo) - Minimal C API example
  • nixos-rk3588 (https://github.com/ryan4yin/nixos-rk3588) - NixOS porting project

Documentation & Learning Resources

  • RKNN SDK Documentation (https://github.com/airockchip/rknn-toolkit2/tree/master/doc) - Conversion and deployment guides
  • Radxa ROCK 5 Documentation (https://docs.radxa.com/rock5) - High-quality English documentation
  • LubanCat Manual (https://doc.embedfire.com/linux/rk3588/quick_start/zh/latest/) - Most comprehensive Chinese documentation
  • RKNPU2 From Beginner to Advanced (Bilibili) (https://www.bilibili.com/video/BV1Kj411D78q) - Video tutorial
  • Rockchip Linux Documentation Repository (https://github.com/rockchip-linux/docs)

Contributing

PRs adding resources are welcome. Please follow these conventions:

  • Resources must be directly related to RK3588 / RK3588S / RK3588J and be accessible
  • Each entry should include [Name](link) - One-sentence description
  • Prioritize actively maintained projects with practical, usable outputs
  • Place under existing categories; new categories can be added if necessary

See CONTRIBUTING.md for details.

License

CC0 1.0 - Public domain contribution, free to use.

Similar Articles

@neil_xbt: https://x.com/neil_xbt/status/2056386359376396458

X AI KOLs Timeline

A comprehensive curated list of 100 repositories across 11 categories that extend Claude Code's capabilities, from awesome lists to memory systems and orchestrators, enabling developers to build a system rather than just use a tool.

@seclink: https://puul.ai: RTL Training Data for Sale to AI Chip Design Labs – 50,000+ Code Bases, 10,000,000+ Raw Files, 10,000+ Reinforcement Learning Environments, Covering Full Project Context (Testbenches, Constraints, Makefiles, Synthesis Scripts, etc.), Organized by Chip Type (TPU-like Systolic Arrays, ASIC Full Flow, etc.).

X AI KOLs Timeline

Puul is selling specialized RTL training data for AI chip design, including 50,000+ code bases and 10,000+ reinforcement learning environments to support AI research in hardware development.