01July 30, 2026/Peter Bhabra

Gigatoken: making tokenisation care about the hardware

A first-principles tour of the SIMD scanners, cache-aware layouts, huge pages and worker scheduling behind Gigatoken.

02July 27, 2026/Jamie Dborin

You Could Have Come Up With Kimi Delta Attention

Guide to the DeltaNet Family of linear attention mechanisms.

03July 23, 2026/Fergus Finn

Throughputmaxxing: DeepSeek-V4-Flash on Isambard-AI

Single-node serving throughput for DeepSeek-V4-Flash: baseline, MLA/DP-attention, and MoE kernel choice.

04July 21, 2026/Fergus Finn

NVLink, NVSwitch, and all that

Why scale-up links are fast and short, how NVSwitch grew the domain from a board to a rack, and what TPU, UALink, and scale-up Ethernet do differently.

05July 9, 2026/Fergus Finn

Reverse-engineering NVIDIA's cuda-checkpoint for faster cold starts

Freezing a live CUDA process to host memory and thawing it again, what the driver does — and doesn't — do to make that work, and how understanding that lets us restore CUDA processes up to 4x faster.

06July 2, 2026/Fergus Finn

Width vs. depth: speculating on the margin

Some thinking about how to trade off batching and speculative decoding in a running inference engine.

07June 30, 2026/Peter Bhabra

The swarm that designs itself

We rebuilt Moonshot's Kimi agent swarm and pointed it at a real codebase: ~53× fewer tokens and ~45× cheaper than one long-context agent.

08June 29, 2026/Fergus Finn

What happens when you run a CUDA kernel

Tracing one vector-add kernel from nvcc all the way down to the warps that execute it.

09June 22, 2026/Jamie Dborin

Prediction: A Frontier Open Source LLM Will Be Released On 3rd December 2026

Using artificial analysis benchmarks we try to predict when an open source LLM will be released that matches frontier LLMs.

10June 22, 2026/Jamie Dborin

Anatomy of a Diffusion Language Model

A breakdown of the modelling choices of three new and popular diffusion language models.

11June 22, 2026/Jamie Dborin

FlashOffload: 7x Cheaper Prefills with Offloading

Extending SGLang to improve it's offloading capabiltiies, achieving speedups for compute bound workloads.

12June 22, 2026/Fergus Finn

Adaptive speculative decoding: picking draft lengths at runtime

A follow-on to the economics of speculative decoding, we run the inference lab simulator on MTP & DFlash drafters with real acceptance data, and find out whether adaptively choosing the draft length is worth it.