MTIA
MTIA (Meta Training & Inference Accelerator)
Meta's own AI silicon
Latest: MTIA 300 deployed; 400/450/500 lineup announced (Mar 11, 2026)
Meta's custom AI silicon line, co-developed with Broadcom: MTIA v1 (2023) and v2 (April 2024) served ranking and recommendations; in March 2026 Meta unveiled four inference chips — MTIA 300 (already live in data centers), 400, 450, and 500 — on a six-month release cadence, targeting generative inference like image and video creation, and reducing reliance on Nvidia.
Why it matters
Meta's path off Nvidia dependence: MTIA v1 and v2 (TSMC 5nm, 708 INT8 TOPS, 90W) served ranking and recommendation models; MTIA 300 is now in production and the 400, 450 and 500 chips target generative inference on a six-month cadence, versus the industry's one-to-two-year cycle, co-developed with Broadcom.
Facts
- MTIA 300 was serving production traffic before the March 2026 announcement.
- A deeper Broadcom co-development partnership was formalized in April 2026.
- The chips deliberately avoid frontier-LLM training — that stays on GPUs.
Try it yourself
March 2026 silicon roadmap (Meta Newsroom) ↗ MTIA v2 deep dive (Meta AI blog) ↗
Sources
Meta AI blog · meta mtia scale ai chips for b ↗Meta AI blog · next generation meta training ↗Meta Newsroom · expanding metas custom silicon ↗Meta Newsroom · meta partners with broadcom to ↗