Zyphra released ZYFA, open-source FP8 and BF16 prefill attention kernels for AMD Instinct MI355X. On real model activations, its FP8 kernels run 1.16-1.76x faster than AMD’s AITER, with 62-78% lower error at head dimension 128.
The post Zyphra ZYFA: FP8 FlashAttention for AMD MI355X, Up to 1.76x Faster appeared first on MarkTechPost.
AMD expects its fourth-quarter revenue to reach roughly $17 billion, potentially surpassing Intel's projected $15 billion to $16 billion and putting AMD ahead of its…
tsv is a set of performance-focused language tools for the Web. Today it ships formatters and parsers for TypeScript/JS, CSS, and Svelte, including a vscode/openvsx…
As anticipated, the AMD PerfOpt feature was picked up for the Linux 7.4 kernel. As shown already in multiple Phoronix benchmarks, AMD PerfOpt makes for a nice…
Discussion (0)
No comments yet. Start the conversation!