BURT CONSULTANTS
Topic

nvidia

3 pieces tagged “nvidia”.

Briefing

Disaggregated inference: lowering token cost for agentic AI.

DigitalOcean’s launch of NVIDIA Dynamo 1.0 claims 67 per cent higher GPU throughput and 79 per cent lower latency. The underlying idea — disaggregated inference — matters for agentic workloads.

nvidiadynamoinferenceagentic ai
3 min read
Briefing

Cache-aware inference routing: how Dynamo cuts LLM latency on AKS.

A Microsoft and Nvidia demonstration shows that KV-cache-aware routing can reduce Time-To-First-Token by around 20x on Azure Kubernetes Service. The result has implications for any team running LLM inference at scale.

akskubernetesnvidiainference
2 min read
Briefing

What Nvidia Vera Rubin means for your AI infrastructure budget.

Nvidia’s Vera Rubin platform promises five times the training compute of Blackwell at a fraction of the token cost. For infrastructure planners, the signal is clear: do not lock in long-term hardware commitments just yet.

nvidiahardwareai infrastructurecost
2 min read

Browse all tags →