# llama.cpp release b11768 adds output index to streaming API

> The open‑source LLM runtime updates its server to include output_index in SSE events and provides new binary builds for macOS, Linux and iOS.

Oossa · 2026-10-21 · https://oossa.com/en/llama-cpp-release-b11768-adds-output-index-to-streaming-api

The ggml‑org/llama.cpp project released version b11768. It fixes the server so that Responses API Server‑Sent Events now include an output_index field, improving compatibility with streaming clients. The change also reuses Anthropic streaming code and cleans up unused tool‑call fields. Pre‑built binaries for macOS (Apple Silicon and Intel), iOS, and several Ubuntu configurations are available for download.

## The facts

- Release version b11768 published on 2026-10-21
- New builds include macOS ARM64, macOS Intel, iOS XCFramework, and Ubuntu x64/arm64 binaries

## Why it matters

Developers using llama.cpp can now reliably track which part of a streamed response each chunk belongs to, simplifying integration with chat applications.

## Sources & references

1. [ggml-org/llama.cpp b11768](https://github.com/ggml-org/llama.cpp/releases/tag/b11768) – llama.cpp, 2026-10-21

Last updated: 2026-10-21