LLM inference can be huge, particularly, with long contexts. In this on-demand video, Junchen Jiang, Assistant Professor at University of Chicago, presents a 10x solution for long contexts inference: an easy-to-deploy stack over multiple vLLM engines with tailored KV-cache backend.
LLM inference can be huge, particularly, with long contexts. In this on-demand video, Junchen Jiang, Assistant Professor at University of Chicago, presents a 10x solution for long contexts inference: an easy-to-deploy stack over multiple vLLM engines with tailored KV-cache backend.
Videos:
Presentation Slides:
Complete the form below to access the full overview:
.png)
Videos
In this talk, Ojus Save walks you through a demo of how to build AI applications on Zoom. This demo shows you an AI agent that receives transcript data from RTMS and then decides if it has to create action items based on the transcripts that are received.
In this talk, Sandeep Joshi, , Senior Manager at NVIDIA, shares how to accelerate the data access between GPU and storage for AI. Sandeep will dive into two options: CPU- initiated GPUDirect Storage and GPU-initiated SCADA.
Bin Fan, VP of Technology at Alluxio, introduces how Alluxio, a software layer transparently sits between application and S3 (or other object stores), provides sub-ms time to first byte (TTFB) solution, with up to 45x lower latency.