inference/research
verifiedhigh · 0.90updated 2026-10-03

Groq LPU AI Inference Technology (architecture overview)

blog

Fields

id
sources/asic-groq-lpu-technology
URL
https://groq.com/lpu-inference-engine
Publisher
Groq
Kind
blog
Published
2026-10-03
Accessed
2026-10-03
notes
Vendor architecture claims: on-chip SRAM bandwidth upwards of 80 TB/s, 14 nm process, no off-chip HBM, deterministic assembly-line execution.

Notes

Vendor architecture claims: on-chip SRAM bandwidth upwards of 80 TB/s, 14 nm process, no off-chip HBM, deterministic assembly-line execution.

Cross-references

This record references no other records.