SYSTEM Cited by 1 source
Meta MTIA (Meta Training and Inference Accelerator)¶
Definition¶
MTIA is Meta's in-house family of AI accelerator silicon — the proprietary-chip axis of Meta's heterogeneous AI hardware fleet alongside NVIDIA GPUs, AMD GPUs, and CPUs. Unlike vendor silicon (NVIDIA / AMD), MTIA "chips are proprietary, no public LLM has been trained on MTIA code" — a forcing function for any code-generation system targeting MTIA to inject the ISA / architecture / memory-hierarchy docs directly rather than rely on pretraining knowledge (Source: sources/2026-04-02-meta-kernelevolve-how-metas-ranking-engineer-agent-optimizes-ai-infrastructure).
Generation cadence — MTIA 300 → 500¶
Meta's MTIA roadmap spans "four chip generations in two years (MTIA 300 through 500), each introducing new compute capabilities, memory bandwidth characteristics, and numeric data types optimized for evolving workloads. A kernel optimized for one generation will underperform when run on the next generation of the same hardware architecture." This two-year cadence is faster than an external silicon vendor's refresh tempo and is the key driver behind Meta needing agentic kernel-authoring systems (KernelEvolve) rather than hand-tuning — no kernel-expert team scales to four generations × every operator × production-model-shape space.
Profiling + instrumentation¶
MTIA's in-house profiling stack — MTIA Insight — exposes accelerator-specific signals not present on GPU platforms:
- PE utilization (processing-element utilization)
- Fixed-function engine metrics (DPE / SFU / MLU utilization and stall cycles)
- Cache behavior
- Per-PE memory bandwidth counters
These structured signals feed KernelEvolve's evaluation harness so the search engine sees "why" a candidate kernel is slow (memory-bound vs compute-bound vs occupancy-limited) — not just a wall-clock speedup number (Source: sources/2026-04-02-meta-kernelevolve-how-metas-ranking-engineer-agent-optimizes-ai-infrastructure).
Programming model¶
MTIA code is written in MTIA C++ (one of the low-level backends KernelEvolve emits kernels for, alongside CUDA for NVIDIA and HIP for AMD). Meta's KernelEvolve generates MTIA kernels via systematic knowledge injection — MTIA architecture manuals, instruction-set references, memory-hierarchy specifications, and optimization patterns are encoded into a retrieval-augmented knowledge base and retrieved on demand when the synthesizer targets MTIA. Canonical wiki instance of hardware proprietary knowledge injection.
Production role¶
On MTIA, KernelEvolve-generated kernels achieve >25% training throughput improvement on an ads model (name not disclosed). The kernels span compute-bound, memory-bound, and custom operations (Source: sources/2026-04-02-meta-kernelevolve-how-metas-ranking-engineer-agent-optimizes-ai-infrastructure).
MTIA is one of the accelerator targets behind Meta Ads's Meta Adaptive Ranking Model serving stack, though the 2026-03-31 MARM post does not itemize per-hardware share.
Seen in¶
- Meta KernelEvolve (2026-04-02, canonical). The proprietary-silicon axis that motivates RAG-over-hardware-docs + agentic kernel synthesis. (Source: sources/2026-04-02-meta-kernelevolve-how-metas-ranking-engineer-agent-optimizes-ai-infrastructure)
Caveats¶
Public information is limited to what Meta discloses on engineering.fb.com and ai.meta.com. Per-generation performance, die size, process node, memory subsystem, and production deployment share are not published. The 2024 MTIA blog post (ai.meta.com/blog/next-generation-meta-training-inference-accelerator-AI-MTIA) and the 2025 MTIA-at-scale post (ai.meta.com/blog/meta-mtia-scale-ai-chips-for-billions) are not ingested on the wiki.
Related¶
- companies/meta — owning company.
- systems/kernelevolve — the agentic kernel-authoring system whose existence is motivated largely by MTIA's proprietary nature + generational cadence.
- systems/meta-adaptive-ranking-model — one of the serving-stack consumers of MTIA-targeted kernels.
- systems/meta-andromeda-ads — the Ads retrieval model; KernelEvolve's MTIA speedup is on an (unnamed) ads model.
- concepts/heterogeneous-ai-accelerator-fleet — the containing forcing function.
- concepts/hardware-proprietary-knowledge-injection — the mechanism that makes MTIA codegen tractable for LLMs that have never seen MTIA code.
- patterns/rag-over-hardware-documentation — the pattern MTIA drove canonicalization of.