Rockchip RKLLM toolkit released for NPU-accelerated large language models on RK3588, RK3588S, RK3576 SoCs

Rockchip RK3588 RKLLM

Rockchip RKLLM toolkit (also known as rknn-llm) is a software stack used to deploy generative AI models to Rockchip RK3588, RK3588S, or RK3576 SoC using the built-in NPU with 6 TOPS of AI performance. We previously tested LLM’s on Rockchip RK3588 SBC using the Mali G610 GPU, and expected NPU support to come soon. A post on X by Orange Pi notified us that the RKLLM software stack had been released and worked on Orange Pi 5 family of single board computers and the Orange Pi CM5 system-on-module. The Orange Pi 5 Pro‘s user manual provides instructions on page 433 of the 616-page document, but Radxa has similar instructions on their wiki explaining how to use RKLLM and deploy LLM to Rockchip RK3588(S) boards. The stable version of the RKNN-LLM was released in May 2024 and currently supports the following models: TinyLLAMA 1.1B Qwen 1.8B Qwen2 0.5B Phi-2 2.7B Phi-3 […]

Linux 6.10 Release – Notable changes, Arm, RISC-V, and MIPS architectures

Linux 6.10 Release Changelog

Linux Torvalds has announced the release of Linux 6.10 on LKML: So the final week was perhaps not quote as quiet as the preceding ones, which I don’t love – but it also wasn’t noisy enough to warrant an extra rc. And much of the noise this last week was bcachefs again (with netfs a close second), so it was all pretty compartmentalized. In fact, about a third of the patch for the last week was filesystem-related (there were also some btrfs latency fixes and other noise), which is unusual, but none of it looks particularly scary. Another third was drivers, and the rest is “random”. Anyway, this obviously means that the merge window for 6.11 opens up tomorrow. Let’s see how that goes, with much of Europe probably making ready for summer vacation. And the shortlog below is – as always – just the last week, not some kind […]

Forlinx FET3562J-C SoM features Rockchip RK3562(J) SoC with 1 TOPS NPU, dual Ethernet, camera, display interfaces

Forlinx FET3562J-C SoM features

The Forlinx FET3562J-C is a Rockchip RK3562/RK3562J powered System-on-Module (SoM) built with a 22nm process and features four high-performance ARM Cortex-A53 cores running at up to 2GHz. It also can support up to 2GB LPDDR4 RAM and can be configured with up to 16GB eMMC storage. Additionally, it has 1 TOPS NPU and standard peripherals like Ethernet, Wifi, Bluetooth, LVDS, MIPI DSI, and much more. All these features make this SoM useful for IoT, automation, and consumer electronic applications. Previously, we have seen Forlinx launch similar SoM with development boards such as the Forlinx RZ/G2L, the FET-MX9352-C, the OK335xS-II and much more feel free to check those out if you are interested in the topic. Forlinx FET3562J-C SoM and carrier board specifications SoM Specifications SoC (one or the other) Rockchip RK3562J – ARM Cortex-A53 quad-core @ 1.8GHz Rockchip RK3562 ARM Cortex-A53 quad-core @ 2.0 GHz NPU – 1 TOPS INT8 […]

Quectel SG368Z WiFi 5 and Bluetooth 4.2 smart module is built around Rockchip RK3568 SoC

Quectel SG368Z Smart Module

The Quectel SG368Z Smart Module is an all-in-one hardware package built around a Rockchip RK3568 AI SoC that combines computing, graphics, storage, and connectivity in a compact form factor. The  LGA module offers WiFi 5 and Bluetooth 4.2 connectivity, dual gigabit Ethernet networking, various video output options (HDMI, LVDS, RGB, MIPI, eDP), USB, PCIe, and many other features making it suitable for applications like smart homes, wearables, and industrial automation. Previously we have seen Quectel introduce various communication modules like the Quectel BG95-S5 5G, the  Quectel KG200Z LoRa the Quectel RG255G RedCap IoT Module, and the Quectel CC660D-LS IoT-NTN module, but this is the first time we have seen Quectel release a Smart Module which can be certainly be considered as a SoM in a LGA package. Quectel SG368Z Smart Module specifications: SoC – Rockchip RK3568 CPU – Quad-core Cortex A55 processor at up to 2.0 GHz GPU – Mali G52 […]

SECO SBC-3.5-RK3568 – A Rockchip RK3568 3.5-inch SBC with dual gigabit Ethernet

SECO SBC 3.5 RK3568

SECO SBC-3.5-RK3568 is a feature-rich 3.5-inch SBC powered by a Rockchip RK3568 quad-core Cortex-A55 AI SoC which includes up to 4GB DDR4-3200 memory, 64GB eMMC 5.1 flash, three display interfaces (HDMI, LVDS, eDP), dual gigabit Ethernet, and various expansion headers for industrial applications. Additionally, it also features, Wi-Fi 802.11 a/b/g/n/ac, Bluetooth 5.0, and LTE support via M.2. USB connectivity includes two USB 3.0 Type-A, and multiple USB 2.0 ports with OTG, alongside RS-232, RS-422, RS-485, and TTL UART ports and more. Previously we have written about similar SBCs powered by the Rockchip RK3568 SoC like the AAEON RICO-3568, the RK3568 Tinker Board 3N, the Radxa ROCK 3B and many others feel free to check those out if you are interested in the topic. SECO SBC-3.5-RK3568 Specifications: SoC – Rockchip RK3568 CPU – Quad-core Cortex A55 processor at up to 2.0 GHz GPU – Mali G52 GPU with support for OpenGL ES 1.1/2.0/3.2, […]

Rockchip RK2118G/RK2118M dual-core Star-SE Armv8-M microcontrollers target smart audio applications

Rockchip RK2118G microcontroller block diagram

Rockchip RK2118G and RK2118M smart audio microcontrollers based on a dual-core Star-SE Armv8-M processor, an NPU for smart AI audio processor, three DSPs, 1024KB SRAM, optional DDR memory in package, and a range of peripherals. I first noticed the RK2118M in slides from the Rockchip Developer Conference 2024 last March, but I did not have enough information for an article at the time. Things have now changed since I’ve just received a bunch of datasheets including the one for the RK2118G and RK2118G microcontrollers, which look identical except for the DDR interface and optional built-in 64MB RAM for the RK2118G. The datasheets have only one reference to Arm with the string “Arm-V8M” and nothing else, and Cortex is not mentioned at all. But the slide above reveals the STAR-SE core looks to be an Arm Cortex-M33 core. We also learn the top frequencies for the “STAR-M33″/”STAR-SE” core  (300MHz) and the […]

Linux 6.9 release – Main changes, Arm, RISC-V, and MIPS architectures

Linux 6.9 release

Linus Torvalds has just announced the release of Linux 6.9 on LKML: So Thorsten is still reporting a few regression fixes that haven’t made it to me yet, but none of them look big or worrisome enough to delay the release for another week. We’ll have to backport them when they get resolved and hit upstream. So 6.9 is now out, and last week has looked quite stable (and the whole release has felt pretty normal). Below is the shortlog for the last week, with the changes mostly being dominated by some driver updates (gpu and networking being the big ones, but “big” is still pretty small, and there’s various other driver noise in there too). Outside of drivers, it’s some filesystem fixes (bcachefs still stands out, but ksmbd shows up too), some late selftest fixes, and some core networking fixes. And I now have a more powerful arm64 machine […]

BIGTREETECH Pi 2 SBC and CB2 module for 3D printers now feature Rockchip RK3566 SoC with Gigabit Ethernet

BIGTREETECH Pi 2 and the BIGTREETECH CB2 Compute Modules

BIGTREETECH recently announced the release of the BIGTREETECH Pi 2 SBC and BIGTREETECH CB2 SoM, both powered by the Rockchip RK3566 SoC. These new modules are direct upgrades from the BIGTREETECH Pi v1.2 and BIGTREETECH CB1, which we covered and reviewed in our previous posts. Specially designed for 3D printers, these new modules have a similar form factor to that of a Raspberry Pi and the RPI CM4 module, and come with a set of upgraded features including support for GbE Ethernet, dual-band WiFi, up to 32GB of LPDDR4 RAM, eMMC storage, HDMI, USB, and more. BIGTREETECH Pi 2 Specifications SoC – Rockchip RK3566 CPU – Quad-core Arm Cortex-A55 processor @ 1.8 GHz GPU – Arm Mali G52-2EE GPU with support for OpenGL ES 1.1/2.0/3.2, Vulkan 1.1, OpenCL 2.0 NPU – 0.8 TOPS AI accelerator VPU – 4Kp60 H.265/H.264/VP9 video decoding, 1080p100f H.265/H.264 video encoding System Memory – 2GB LPDDR4 (Customizable – 1GB/2GB/4GB/8GB) […]

Exit mobile version
EmbeddedTS embedded systems design