Active params in L2 cache hit 14,572 tok/s on Intel Emerald Rapids with AMX
GregoryDiamos · x · 2026-09-07
Gregory Diamos needed a 10k tok/s CPU model for data processing, so he gave Claude Code a pile of tokens to build an outrageously small neural net. With active params fitted into L2 cache on an Intel Emerald Rapids CPU with AMX, it reaches 14,572 tok/s. He shares three interesting discoveries and argues tiny nets deserve a revisit for inference/data-processing workloads.
More from Infra
- A100 rental prices still rising in 2026 per a16z data, Jevons paradox hits compute — VraserX · 2026-10-27
- Anthropic S-1 reveals $800B+ in off-books future compute commitments — GaryMarcus · 2026-10-27
- ~42% of Anthropic's ~11GW contracted compute is NVIDIA GPUs — ns123abc · 2026-10-27
- EdgeRock-4B holds KV cache at 72 MiB across 256K context, 99.1% smaller than Qwen3.5-4B — alifcoder · 2026-10-27
- RTX 5080 price jumps from $1658 to $2318 in ten days on Amazon — Frizzy-MacDrizzle · 2026-10-27
- ComfyUI on 12GB VRAM: measured peaks show offloading turns 24s into 11 minutes — Impossible-Mail-4619 · 2026-10-27