# llama.cpp adds LFM2 tiling support and expands platform builds

> The open‑source LLM runtime now reads tiling parameters from GGUF files and ships binaries for many new CPU, GPU and OS targets.

Oossa · 2026-10-21 · https://oossa.com/en/llama-cpp-adds-lfm2-tiling-support-and-expands-platform-builds

The ggml‑org team updated llama.cpp to read LFM2‑VL tiling settings from GGUF metadata. This lets the model split images into tiles based on min_tiles, max_tiles, tile_size and max_pixels_tolerance values stored in the model file. The release also bundles pre‑compiled binaries for a wide range of platforms – macOS, Linux, Windows, Android and various hardware accelerators such as CUDA 13, Vulkan and OpenCL.

## The facts

- Release tag b11767 published on Wed Oct 21 2026
- Adds support for reading LFM2 tiling params from GGUF metadata

## Why it matters

Developers can now run image‑processing LLMs on more devices without writing custom tiling code.

## Sources & references

1. [ggml-org/llama.cpp b11767](https://github.com/ggml-org/llama.cpp/releases/tag/b11767) – llama.cpp, 2026-10-21

Last updated: 2026-10-21