PrefixBench-H100: Characterizing Prefix Reuse and Time-to-First-Token in H100 LLM Serving
## Episode Summary In this episode, we cover: - **PrefixBench-H100: Characterizing Prefix Reuse and Time-to-First-Token in H100 LLM Serving** (arXiv) - **Understanding and Exploiting Diagonal Attention Sparsity in Autoregressive Image Generation** (arXiv) - **Checking In On The ISA Wars And Its Impact On CPU Architectures - Hackaday** (google_arch) - **Kari Hepola: Custom tools make RISC-V processors faster and easier to design - tuni.fi** (google_riscv) - **Enabling Comprehensive CWE-based Assurance for RISC-V Processors - semiengineering.com** (google_riscv) --- *Sponsored by Ada, Ago…
The skinny
The skinny isn't ready yet — notes appear once the transcript is processed.