# llama.cpp b11532 pre‑release adds exact GELU for ModernBERT

> The open‑source Llama.cpp library released version b11532 on Oct 9 2026, improving ModernBERT encoder accuracy.

Oossa · 2026-10-09 · https://oossa.com/en/llama-cpp-b11532-pre-release-adds-exact-gelu-for-modernbert

The ggml‑org team published a pre‑release of llama.cpp labeled b11532. The update adds exact GELU activation for ModernBERT encoders and keeps tanh‑GELU aliases. It also maps the Python GELU function to a more precise implementation. The release includes builds for many platforms, from macOS Apple Silicon to Windows CUDA 13.4. The source code is signed with a verified GPG key.

## The facts

- Release tag b11532 on 2026‑10‑09
- Adds exact GELU for ModernBERT encoders

## Why it matters

Developers can run ModernBERT models locally with more accurate activation functions, improving inference quality without cloud services.

## Sources & references

1. [ggml-org/llama.cpp b11532](https://github.com/ggml-org/llama.cpp/releases/tag/b11532) – llama.cpp, 2026-10-09

Last updated: 2026-10-09