# llama.cpp adds H2O‑Lightning‑4B decision model

> The open‑source inference library now supports the H2O‑Lightning‑4B model, a Qwen3.5‑4B with a decision‑type LoRA.

Oossa · 2026-10-21 · https://oossa.com/en/llama-cpp-adds-h2o-lightning-4b-decision-model

The ggml‑org/llama.cpp project released version b11773 on 2026‑10‑21. The update adds a new model type called H2O‑Lightning‑4B, which is a Qwen3.5‑4B model with a merged decision LoRA. The model returns answers prefixed with a space (" A") and can handle up to 255 labels. The converter now writes decision metadata and checks the serve_config.json for a match. The change was co‑authored by H2O AI staff and tested on a range of CPU and GPU platforms.

## The facts

- Model: H2O‑Lightning‑4B decision type
- Release tag b11773 published 2026‑10‑21

## Why it matters

Developers can now run decision‑type AI models directly with llama.cpp on many hardware setups.

## Sources & references

1. [ggml-org/llama.cpp b11773](https://github.com/ggml-org/llama.cpp/releases/tag/b11773) – llama.cpp, 2026-10-21

Last updated: 2026-10-21