GGML | AIlisting.ai

GGML

(0)

★★★★★ 0.0/5

(0)

★★★★★ 0.0/5

Home > Ai Tools > Code > Code Assistant > GGML

What is GGML?

GGML is a machine learning tensor library designed for large models and high performance, featuring 16-bit float support and more.

Free

Open Source

Social Links

GGML Features

Written in C
Supports 16-bit floats and integer quantization (e.g., 4-bit, 5-bit, 8-bit)
Automatic differentiation
Built-in optimization algorithms (e.g., ADAM, L-BFGS)
Optimized for Apple Silicon and x86 architectures
Web support via WebAssembly and WASM SIMD
No third-party dependencies
Zero memory allocations during runtime
Guided language output support
Open source under the MIT license

Use Cases of GGML

whisper.cpp: High-performance inference of OpenAI's Whisper automatic speech recognition model.
llama.cpp: Inference of Meta's LLaMA large language model.

Contributing: You can support the project by contributing to the codebase or becoming a sponsor. The development process is open, and innovative ideas are encouraged.

Company: GGML.ai, founded by Georgi Gerganov, supports GGML's development, with sponsorship from Nat Friedman and Daniel Gross. They are seeking full-time developers to advance on-device inference.

GGML Reviews

Please login to write a review.

Similar Tools

Sketch

Sketch is an AI code-writing assistant for pandas users, making code suggestions with context-awaren...

Reflect.run

Refact is an AI coding assistant that enhances programming by suggesting code completions, refactori...

PseudoEditor

The Pseudocode Online Editor is a fast, free, and web-based tool that simplifies pseudocode writing ...