Other

whisper.cpp README

Georgi Gerganov, GitHub, 2026

Read the original

Read on October 16, 2026

What it is: The README of whisper.cpp, a C and C++ implementation of OpenAI's Whisper speech recognition model that runs on CPUs and several kinds of GPU.
How much weight it carries: Primary project documentation, widely used as the engine inside desktop dictation apps. It lists capabilities, not benchmarks.

Facts taken from this source

The first run of whisper.cpp on a device with Core ML is slow, because the Neural Engine service compiles the model to a device-specific format.
“The first run on a device is slow, since the ANE service compiles the Core ML model to some device-specific format.” (Core ML support)
The large model in whisper.cpp needs about 3.9 GB of memory and takes 2.9 GiB on disk.
“| large | 2.9 GiB | ~3.9 GB |” (Memory usage)
Quantized whisper.cpp models require less memory and disk space, and can run more efficiently on some hardware.
“Quantized models require less memory and disk space and depending on the hardware can be processed more efficiently.” (Quantization)
The whisper.cpp project describes Apple silicon as a first-class target for running OpenAI's Whisper model, optimized through ARM NEON, the Accelerate framework and Metal.
“Apple Silicon first-class citizen - optimized via ARM NEON, Accelerate framework, Metal and” (Feature list)
The whisper.cpp command-line example runs only with 16-bit WAV files, so audio in other formats must be converted before transcription.
“currently runs only with 16-bit WAV files, so make sure to convert your input before running the tool.” (Usage (README))
The whisper.cpp project shows Whisper running fully offline and on the device on an iPhone 13.
“here is a video of running the model on an iPhone 13 device - fully offline, on-device” (Introduction, demonstration of whisper.objc)
The whisper.cpp README describes the project as high-performance inference of OpenAI's Whisper automatic speech recognition model.
“High-performance inference of [OpenAI's Whisper](https://github.com/openai/whisper) automatic speech recognition (ASR) model:” (Introduction)
The whisper.cpp project says that running the Whisper encoder on the Apple Neural Engine through Core ML can be more than three times faster than CPU-only execution on Apple silicon.
“This can result in significant speed-up - more than x3 faster compared with CPU-only execution” (Core ML support section)
The whisper.cpp README says that with a dedicated NPU, it can run the Whisper encoder on the NPU through VitisAI, which gives a significant speedup over CPU-only inference.
“with a dedicated NPU, whisper.cpp can fully offload the Whisper encoder to the NPU via VitisAI, delivering significant speedup over CPU-only inference” (README, NPU section (VitisAI))
The whisper.cpp project says its quantized Whisper models require less memory and disk space, and that depending on the hardware they can be processed more efficiently.
“Quantized models require less memory and disk space and depending on the hardware can be processed more efficiently.” (Quantization section)
The whisper.cpp project lists support for CPU-only inference among its features, alongside GPU support for NVIDIA, AMD and Vulkan.
“Support for CPU-only inference” (Features)

whisper.cpp README

Facts taken from this source

Pages that rest on this source

whisper.cpp README

Facts taken from this source

Pages that rest on this source