Tag: #amd

AMD publishes detailed Instella‑MoE technical report; weights licensed for research use

AMD published a detailed technical report and checkpoints for Instella‑MoE, a 16B MoE model trained on its Instinct GPUs — but model weights are licensed for research only.

$AMD, #amd, #ai, #moe, #openmodels

AMD Says Sixth‑Gen EPYC 'Venice' on Track for Launch Later in 2026

AMD said on its Q1 2026 call that sixth‑gen EPYC 'Venice' (Zen 6, TSMC 2nm, up to 256 cores) remains on track to launch later in 2026.

$AMD, #amd, #epyc, #server, #processors, #tsmc