# llama.cpp b11757 release adds CUDA graph cleanup

> The open‑source Llama inference library published version b11757 with a fix for CUDA graph handling and new binaries for macOS, iOS and Linux.

Oossa · 2026-10-20 · https://oossa.com/en/llama-cpp-b11757-release-adds-cuda-graph-cleanup

The ggml‑org team released llama.cpp version b11757. The update clears CUDA graphs after a legacy memory‑pool flush, which should reduce GPU memory leaks. Pre‑built binaries are now available for Apple Silicon macOS, Intel macOS, iOS, and several Ubuntu builds (CPU, Vulkan and CUDA 12.8). The release is posted on GitHub.

## The facts

- Release tag: b11757
- Release date: 2026‑10‑20

## Why it matters

Developers running Llama models on Nvidia GPUs will see fewer out‑of‑memory errors after the fix.

## Sources & references

1. [ggml-org/llama.cpp b11757](https://github.com/ggml-org/llama.cpp/releases/tag/b11757) – llama.cpp, 2026-10-20

Last updated: 2026-10-20