Delivering high-quality AI-powered applications historically relied on massive models. That came with significant scaling limitations, as deploying models with more than 100B parameters and maximizing toke generation doesn’t scale up without losing latency… Read More
Conversations around fast inference typically focus on one approach: blazing fast token generation with gargantuan models. Both still play an important part in many circles. Models come in huge flavors, including Qwen (235B)… Read More
The first cloud computing revolution triggered a tsunami-grade buildout for data centers—both for emerging hyperscalers like Amazon Web Services, and companies re-orienting their systems around cloud-based operations. At the time, we were… Read More
Today, some of the most prevalent AI-powered applications are highly interactive ones: coding companions, customer service chatbots, and others. While they all have different measures of success, they’re increasingly aligned on one… Read More
In the early days of transformers-based AI workloads, generation speed wasn’t that big of an issue. Users generally tolerated slower inference due to novelty, and serious use cases outside of… Read More
Time-to-first-token, or TTFT, has generally been one of many key performance indicators for AI-based applications. Consumer-facing chatbots, for example, might need it to be very low, while enterprise cases can… Read More
d-Matrix’s advantage over classic GPU architecture is extremely low latency, ultra-efficient small batch inference. We approach each of those problems with a large range of techniques and must excel at… Read More
In today’s rapidly evolving AI landscape, it’s clear that inference — not just training — is becoming the new scaling challenge. As models grow in size and capability, the infrastructure… Read More
We are inevitably moving toward a world where AI applications need substantially better performance, and classic GPUs alone can’t keep up. And at the same time, customers need flexibility in… Read More
Large-scale AI workloads are deployed by stringing together many cards and many nodes. Any increase in model size, context length or users impacts the number of cards needed. Gone are… Read More
