So the big news today is how the guys in China finally did another Top500 submission, at #1. To be clear - China has had a ~Top10 HPC system for the past few years - the upgraded TaihuLight - but has refrained from making a submission for fear of reprisals in the form of sanctions. This is probably a good idea - nation-scale systems are unashamedly used for weapons design and flaunting this capability in front of the world when you are not using fully domestic technology will generally lead to a bad ending. The new submission - LineShine - achieves 2.2EFLOPS on HPL using 45360 Huawei LX2 CPUs connected using a proprietary fat-tree network. The interesting thing here is the LX2 CPU, which is the latest in a series of Chinese HPC processors built with a explicitly-memory-managed, DSP-like architecture. We briefly dive into the architecture below, as well as some historical comparisons. All HPC devices have a shared memory-global memory architecture, where shared memory is a fast local SRAM accessible by a subset of the processing units and global memory is DRAM which is accessible by all processing units. Where they differ is how this hierarchy is managed: - CPUs use the shared memory as cache, hiding it entirely from the programmer - GPUs use the shared memory as cache, but additionally allows explicit shared memory loads - DSPs only allow computation on shared memory, relying on explicit loads to transfer data from global memory Additionally, CPU/DSPs and GPUs differ in how they expose parallelism to programmers: - CPU/DSPs explicitly reveal cores and SIMD lanes: there are N cores, each of which an execute m operations per cycle - GPUs abstract cores x SIMD lanes as threads: to the programmer, there are N threads, each of which work on one element at a time. Internally, GPUs are a complex share-resource SIMD multicore processor - GPU "cores" are not fully independent of each other, but are more so than the SIMD lanes in a CPU The LineShine/Huawei LX2 is a 304-core ARM DSP which is physically two 152-core processors and electrically 8 38-core core groups. Each core group has access to 4 GB of HBM and each 152-core processor has access to 128 GB of DDR5. While it appears that cores do have access to memory directly via load/store instructions (as an ARMv9-compliant device, it would be impossible not to), the primary way of memory access is through the DMA engines, a block which lives in the uncore and asynchronously moves data between memory levels. The most famous DMA-driven processor was the Cell Broadband Engine, a 1 8 core DSP where the 8 cores could only access data via DMA transfers. Its spiritual successor is the Sunway SW26010P, a 384-core processor built as 6 64-core complexes, each in the same style as Cell. LX2 is a divergence from the pure DMA architecture - there is no management core - but the idea remains the same: explicit programmer management of data transfer leads to high performance in the hands of a skilled programmer.

The First 100 is live. We partnered with 11 of the best teams in crypto and AI to get here. They tested the platform, scored real robot data, and helped shape the standard before launch. Meet our Day 1 Launch Partners.

The First 100 begins now. Verify Quality is live on PrismaX. For the first time, anyone can score the robot training data that models learn from, earn points, and compete to become one of The First 100. Better data. Better models. The standard starts with you.

huge congrats to the awesome team at @ambient_xyz ! even if you don't care about the philosophy of open access, here's why what they're doing is important: post-2026 LLM workflows rely not just on LLM access, but consistent, repeatable behavior from the same model, call after call, month after month. we're not talking about some customer service chatbot that sits on your webpage doing nothing, but serious, internal tooling designed to act like a junior developer or junior marketing associate, taking over repetitive tasks so that your humans can focus on being productive human beings. these tools need to be well-characterized - it's honestly OK if they make mistakes once in a while, but their users need to know what mistakes can happen and how often in order to make the correct decisions. random infrastructure changes that may feel benign or even beneficial for run-of-the-mill chat users can wreak havoc on carefully-tuned agentic pipelines. this brings me to my point: in tech, nothing lasts forever. it may feel unthinkable that Anthropic, or Fireworks, or Google, will be gone someday, but if you look at this industry's history the only proprietary closed-source standard that has survived until 2026 is the Microsoft Windows API. if you built your company on HP-UX, or SGI, or Netscape, or DEC Alpha, you'd be out of luck today - either the standards or the companies themselves have ridden off into the sunset. Ambient is a huge step in the right direction. even if all the organizations named in this post vanished today, Ambient could still live on. being a proper, permissionless system means that as long as there is a single person interested in the project, the project can be revived. projects like Ambient are a vital part of solving a hard problem: how do we bring the same kind of decentralized resilience that the human-generated Internet enjoys to knowledge stored in language models, while managing the fact that unlike the Internet, creating and using these models consumes tremendous physical resources?

It's amazing how quickly technology progresses when there's money to be made. About a decade ago the state of the art in rack-scale interconnects was NUMALINK 7, which dated back to the SGI of the mid-90s and was used by weird scientific computing people for their aero/weather/nuclear simulations. The total number of customers was about 10 (SGI would put out a press release every time they sold a system). NUMALINK 7 was good for about 100 gbits/second per cable. The spiritual successor to NUMALINK is definitely rack-scale NVLINK, which is similarly a copper-based, performance-oriented, short-range interconnect. NVLINK 5 is good for an incredible 14400 gbits/sec - over two orders of magnitude faster than NUMALINK in 10 years. And yes, people complain about the cost of NVL72 systems, but a rack of Blackwell in 2026 costs less than a rack of SGI in 2016...

what a curious contraption. the actuators are a series of fluid-filled pouches which puff up and shorten when voltage is applied. these pull on cables which actuate the fingers. very cool, but I'm sure MTBF is highly suspect...

finally, we get onshore actuators. congrats on the round and much thanks to a16z for helping give the industry a much needed building block

11/N and finally, a shot of the show floor. not as cool as IROS Hangzhou but some new faces for sure

10/N misc photos of possible interest

9/N this panda is a dog in a suit. and yes, you can buy it

8/N this biped called itself the Kangaroo and had a pretty unique screw-driven hip joint with no distal motors. even the ankle motors are at the hip

7/N even the drones are coming back. the flapping one was pretty cool - the photo was taken at 1/4000 sec shutter speed to freeze the bending of the wings

6/N meanwhile, we get this gem of a slogan from the mighty Unitree...

5/N Sharpa was showing off by having one of their robots hand out brochures

4/N the autonomous toy car competition was impressive. these cars were whizzing around the track too fast to see

3/N did you know Nao is alive and well? coming up on version 7, too

2/N a handful of Chinese hardware companies had working VLAs (probably Pi finetunes) on their hardware

Assorted ICRA '26 commentary 1/N: HiTorque robotics brought a fleet of Mini Pi. I really like these robots: strong compute, open architecture, and stubby form factor make them great for locomotion research

some of my model params accidentally got upcast into fp32 during a refactor. the subsequent hunt for the performance regression was so arduous that Claude Opus emitted things like ?? and an all caps "🎯 SMOKING GUN FOUND" complete with emoji