LocalAI

LocalAI is an open-source AI engine that runs LLMs, vision, voice, image, and video models locally.

LocalAI is an open-source AI engine that lets users run text, voice, vision, image, video, 3D, and agent workloads on their own hardware. It provides a drop-in replacement with an OpenAI-compatible API, and also supports Anthropic, Ollama, and ElevenLabs APIs so existing clients only need a URL change. The runtime pulls in swappable backends like llama.cpp, vLLM, SGLang, and MLX dynamically as models request them. Every feature ships a CPU path first, allowing users to run models without requiring a GPU. It is designed for developers and organizations wanting to keep data on their own infrastructure, scaling from a single laptop to a distributed cluster.

LocalAI at a glance

Best for: Software engineer, Researcher
Runs on: self-hosted, open source, API, CLI
Pricing: Free
Released: March 2023
Category: AI coding tool
Website: localai.io

Key features

OpenAI-compatible API with swappable backends
Realtime voice API with WebRTC and streaming ASR
Live transcription with speaker labels and timestamps
Sound event detection for 527 kinds of audio
Voice and face recognition with liveness detection
3D reconstruction and distance estimation from photos
APEX quantization for fitting large models into smaller memory
Distributed mode with smart routing and autoscaling

Use it for

Running LLMs and vision models locally without a GPU
Building real-time voice applications with WebRTC
Transcribing meetings with speaker diarization
Executing local AI agents with shell command access
Cloning voices and generating speech across multiple languages

Alternatives to LocalAI

Portable Local Studio — Runs completely offline with zero configuration and automatic hardware acceleration.
WebAI by Opera — Runs local open-weight models directly in your browser without cloud costs.
Open WebUI — Run AI on your own terms with complete data privacy and self-hosting control.
Local Code Community — Runs local offline models and built-in tools securely inside an app sandbox.
Ollama — Run open models locally or in the cloud with private prompts
Local Deep Research — Fully local and encrypted execution with proper citations.
AssemblyAI — Industry-leading speech models deliver high accuracy and low latency for voice applications.
Lemonade — It provides private, local AI execution with OpenAI API standard support.
Foundry Local — Runs AI models locally on device with complete privacy and no Azure subscription.
Project Maya — Runs a 321B-parameter AI model locally across VRAM, RAM, and NVMe SSD.
Strata — Runs large AI models locally on consumer gaming PCs without server reliance.
Text Generation WebUI — Run local LLMs with multiple backends, multimodal input, and tool calling 100% offline.

All LocalAI alternatives