ggml-org/llama.cpp

ggml : add ANE backend

Open

#10,453 opened on Nov 22, 2024

View on GitHub
 (21 comments) (84 reactions) (0 assignees)C++ (110,169 stars) (18,202 forks)batch import
help wantedresearch 🔬roadmap

Description

According to this https://github.com/ggerganov/llama.cpp/discussions/336#discussioncomment-11184134, there is a new CoreML API and an ANE backend might be possible to implement with latest Apple software/hardware.

Contributor guide

Tech stack
cpp
Domain
backendmachine learningperformance
Issue type
feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
5
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
over 1 week
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
mostly clear
Prerequisites
C++ggmlCoreMLApple Neural Engine
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
5
Research direction
Review the referenced discussion comment and Apple's CoreML documentation. Investigate the existing Metal backend in ggml for integration patterns. Explore the ANE capabilities via CoreML Performance Shaders. Identify required API calls and model conversion steps for ggml.