metaai·lightalo unofficial · independent
Universe / On-Device & Silicon / Meta LLM Compiler
On-Device & Silicon · 2024

Meta LLM Compiler

77% of a full autotuning search's gains, without compiling once

open source 7B / 13B (plus FTD variants) params

Latest: LLM Compiler 7B/13B + 7B-ftd/13B-ftd (Jun 27, 2024)

Foundation models (7B/13B, plus fine-tuned FTD variants) built on Code Llama for compiler optimization: trained on 546B tokens of LLVM-IR and x86/ARM/CUDA assembly to emulate the compiler, tune optimization flags, and disassemble binaries back to IR. Released June 27, 2024 under the bespoke Meta LLM Compiler License permitting both research and commercial use.

Why it matters

The first foundation models for compiler optimization: trained on 546B tokens of LLVM-IR and assembly, the 13B model emulates compiler optimizations 20% of the time versus Code Llama's 0.8%, and the FTD variants improve code size by 4.88% and disassemble with 0.96 round-trip BLEU. Released under a license permitting commercial use.

Facts

Try it yourself

Lineage

Descends fromCode Llama

See the whole family tree →

Sources

More in On-Device & Silicon

MobileLLMGPU superclusters: 350K H100s → Prometheus & HyperionQuantized Llama 3.2MTIAMobileLLM-R1 / R1.5MobileLLM-Pro

Read the On-Device & Silicon story on the sky →

✦ Open on the map Explore On-Device & Silicon Quiz me