Skip to main content
PromptQuorum
Home/Power Local LLM/The Complete Local LLM Software Directory: 130 Tools to Run AI on Your Own Hardware (2026)
Overview & Reference

The Complete Local LLM Software Directory: 130 Tools to Run AI on Your Own Hardware (2026)

·28 min read·By Hans Kuepper · Founder of PromptQuorum, multi-model AI dispatch tool · PromptQuorum

130 local LLM tools across 7 categories, each with its licence, price, and primary URL. Use the filters below to find the right one for your stack.

130 local LLM tools across 7 categories — Run & Serve, Chat & Assistants, Code & Development, Knowledge & Retrieval, Voice & Audio, Images & Video, and Train & Operate. Filter, search, and compare below.

151 tools

Fully local: 76Hybrid: 75Cloud: 0

LM Studio

General chat clients
100% local

Most polished GUI, built-in HuggingFace model browser, server mode

Runs its own engineFree
≈8 GB RAM for a 7B model
Desktop appmacOSWindowsLinux
107 articles
Closed source

Atomic Chat

General chat clients
100% local

Offline desktop and mobile chat app with one-click local agents

Runs its own engineFree
≈8 GB RAM for a 7B model
Mobile appiOSAndroid
3 articles
Apache 2.0

Msty

General chat clients
Hybrid

Clean consumer UX, multi-provider support

Own engine + externalFree + paid tier
OllamaLM Studiollama.cppOpenAI API
≈8 GB RAM for a 7B model
Desktop appmacOSWindowsLinux
10 articles
Closed source

Backyard AI

Roleplay & companions
100% local

Character chat and roleplay desktop client

Runs its own engineFree + paid tier
≈8 GB RAM for a 7B model
Desktop appmacOSWindows
3 articles
Closed source

BoltAI

General chat clients
Hybrid

Native macOS desktop AI client with Ollama support

Own engine + externalFree + paid tier
OllamaLM Studio
≈8 GB RAM for a 7B model
Desktop appmacOS
6 articles
Closed source

Draw Things

Image generation
100% local

Local image generation on macOS and iOS with Stable Diffusion

Runs its own engineFree
≈8 GB RAM for a 7B model
Desktop appMobile appmacOSiOS
1 article
Closed source

Hanoki

General chat clients
Hybrid

macOS branching chat, local via Ollama or cloud APIs

Own engine + externalFree + paid tier
OllamaOpenAI APIAnthropic APIGoogle Gemini APIOpenRouter
≈8 GB RAM for a 7B model
Desktop appmacOS
1 article
MIT

Open Felix

Autonomous agents
Hybrid

macOS agent, local on Apple Silicon or escalates to cloud

Own engine + externalFree + paid tier
≈8 GB RAM for a 7B model
Desktop appmacOS
1 article
Apache 2.0

Osaurus

General chat clients
100% local

Native macOS app, local models via Ollama/MLX/LM Studio

Own engine + externalFree
OllamaLM Studio
≈8 GB RAM for a 7B model
Desktop appmacOS
1 article
MIT

BoBe

General chat clients
100% local

Free local AI desktop assistant for Mac, on-device

Runs its own engineFree
≈8 GB RAM for a 7B model
Desktop appmacOS
1 article
Open source

Voxa

Real-time voice agents
Hybrid

Local-first desktop voice assistant, optional cloud voice

Own engine + externalFree + paid tier
≈8 GB RAM for a 7B model
Desktop appmacOSWindows
1 article
MIT

Jarvis

Real-time voice agents
100% local

macOS voice assistant, fully offline (Llama, Whisper, Kokoro)

Runs its own engineFree
≈8 GB RAM for a 7B model
Desktop appmacOS
3 articles
Open source

Windsurf

Code assistants & IDE plugins
Hybrid

AI-first IDE with local model integration

Needs Ollama/LM StudioFree + paid tier
Set by your engine
Desktop appmacOSWindowsLinux
8 articles
Closed source

Sourcegraph Cody

Code assistants & IDE plugins
Hybrid

AI code assistant with local model support

Needs Ollama/LM StudioFree + paid tier
Ollama
Set by your engine
Editor / notes pluginmacOSWindowsLinux
4 articles
Apache 2.0

CodeGPT

Code assistants & IDE plugins
Hybrid

IDE integrations across multiple editors

Needs Ollama/LM StudioFree + paid tier
OllamaLM Studio
Set by your engine
Editor / notes pluginmacOSWindowsLinux
1 article
MIT

Cursor (local mode)

Code assistants & IDE plugins
Hybrid

AI-first code editor with local model support

Needs Ollama/LM StudioFree + paid tier
Set by your engine
Desktop appmacOSWindowsLinux
Closed source
Get it ↗

Bodega One Code

Code assistants & IDE plugins
Hybrid

Local-first AI coding IDE, BYOLLM via Ollama/LM Studio

Needs Ollama/LM StudioFree
OllamaLM Studiollama.cppLocalAIKoboldCpp
Set by your engine
Editor / notes pluginmacOSWindowsLinux
3 articles
Closed source

Blackbox AI (CLI)

Code assistants & IDE plugins
Hybrid

Terminal code generation and chat from your shell

Needs Ollama/LM StudioFree + paid tier
Set by your engine
Command linemacOSWindowsLinux
Apache 2.0
Get it ↗

Hermes Agent

Autonomous agents
Hybrid

Self-improving personal AI agent, persistent memory

Needs Ollama/LM StudioFree
OllamaLM StudiovLLMllama.cpp
Set by your engine
Command linemacOSLinux
1 article
MIT

Msty Go

Autonomous agents
Hybrid

Autonomous multi-step task agent, local or cloud models

Own engine + externalFree + paid tier
Ollamallama.cpp
≈8 GB RAM for a 7B model
Desktop appmacOSWindowsLinux
3 articles
Closed source

XTTS v2

Voice cloning
100% local

Voice cloning from short audio samples with multilingual support

Library / SDKFree
Set by the model you load
Library / SDKmacOSWindowsLinux
8 articles
CPML

Ollama vision models

Vision & OCR
100% local

Vision-capable models via Ollama (Llama 3.2 Vision, LLaVA, etc.)

Needs Ollama/LM StudioFree
Ollama
Set by your engine
Command linemacOSWindowsLinux
4 articles
Various

Idefics

Vision & OCR
100% local

Open-source multimodal model for vision and language

Library / SDKFree
Set by the model you load
Library / SDKmacOSWindowsLinux
2 articles
Apache 2.0

Private LLM

General chat clients
100% local

Polished iOS and macOS local LLM app

Runs its own enginePaid
≈8 GB RAM for a 7B model
Mobile appiOSmacOS
14 articles
Closed source

This page contains links to third-party products for reference. PromptQuorum is not enrolled in any affiliate program — these are plain links that earn no commission. Clicking links and your next steps are entirely your own responsibility. These links do not represent any endorsement or verification by PromptQuorum.

Common Real-World Stacks

For readers who do not want to read all seven categories, pick the closest stack and copy it. Each row pairs a real goal with a tested combination and the hardware floor it actually runs on.

Goal
Stack
Hardware floor
Just chat casuallyLM Studio standalone16 GB RAM, no GPU
Best balance for power usersOllama + Open WebUI16 GB RAM, optional GPU
Document chatOllama + AnythingLLM16 GB RAM, optional GPU
CodingOllama + Continue.dev16 GB RAM + GPU recommended
Roleplay / creativeKoboldCpp + SillyTavern16 GB RAM, GPU recommended
Privacy-first businessOllama + Open WebUI + PrivateGPT32 GB RAM + 12 GB VRAM
Mobile / on-the-goMLC Chat or PocketPal AIiPhone 13+ / Pixel 7+
Apple SiliconOllama (MLX backend) or LM StudioM2/M3/M4/M5 with 16+ GB unified
Multi-user teamvLLM + Open WebUI32+ GB RAM + multi-GPU
Image generationStable Diffusion + ComfyUI or Invoke AI6+ GB VRAM GPU
Voice assistantOllama + Whisper.cpp + Piper TTS8 GB RAM, CPU-only possible
10+ common real-world local AI stacks by goal: from LM Studio standalone (16 GB RAM, no GPU) to vLLM + Open WebUI for multi-user teams (32 GB RAM + multi-GPU), Stable Diffusion for images (6 GB VRAM), and Ollama + Whisper + Piper for fully-offline voice assistants. Ollama + Open WebUI is the best-balance default at 16 GB RAM.
10+ common real-world local AI stacks by goal: from LM Studio standalone (16 GB RAM, no GPU) to vLLM + Open WebUI for multi-user teams (32 GB RAM + multi-GPU), Stable Diffusion for images (6 GB VRAM), and Ollama + Whisper + Piper for fully-offline voice assistants. Ollama + Open WebUI is the best-balance default at 16 GB RAM.

Key Takeaways

  • Seven categories, 130 projects, one map. Run & Serve, Chat & Assistants, Code & Development, Knowledge & Retrieval, Voice & Audio, Images & Video, and Train & Operate — most popular projects in 2026 fit in one primary category, and many span more than one.
  • Pick a runtime first. Ollama is the right default for ~95% of readers; llama.cpp is the foundational engine underneath most other tools; vLLM is the production-serving pick for multi-user setups.
  • Most categories above Run & Serve are optional. A desktop app OR a web UI is enough for chat. Add a code assistant or CLI tool only when you want code help; add a RAG system only when you want to chat with your own documents; add an agent framework only when one-shot calls stop being enough; add Images & Video only when you need visual output.
  • Licence matters for commercial use. MIT and Apache 2.0 dominate the ecosystem. AGPL appears on a handful of UIs (text-generation-webui, KoboldCpp, Jan, SillyTavern) — fine for personal use, more deliberate for commercial deployments. The "License" column below names every one explicitly; see AI & open-source software licenses explained for what MIT, Apache-2.0, AGPL-3.0, and the other license types below actually require.
  • Multi-tool stacks are normal. Ollama + Open WebUI + AnythingLLM + Continue.dev + Stable Diffusion is a single-machine setup that covers chat, RAG, coding, and image generation without compromise. The "Common Real-World Stacks" table below names the recipes that actually work in 2026.
The 7 categories of a local LLM stack: 130 actively-maintained projects spanning Run & Serve (Ollama, llama.cpp, vLLM), Chat & Assistants (LM Studio, Jan, GPT4All), Code & Development (Continue.dev, LangChain, CrewAI), Knowledge & Retrieval (AnythingLLM, PrivateGPT), Voice & Audio (Whisper.cpp, Piper), Images & Video (Stable Diffusion, ComfyUI), and Train & Operate.
The 7 categories of a local LLM stack: 130 actively-maintained projects spanning Run & Serve (Ollama, llama.cpp, vLLM), Chat & Assistants (LM Studio, Jan, GPT4All), Code & Development (Continue.dev, LangChain, CrewAI), Knowledge & Retrieval (AnythingLLM, PrivateGPT), Voice & Audio (Whisper.cpp, Piper), Images & Video (Stable Diffusion, ComfyUI), and Train & Operate.

How This Directory Stays Current

This directory is reviewed every six months and patched between reviews. Recent expansions added dozens of new tools across every category, split voice/multimodal into three focused categories (speech-to-text, text-to-speech, vision), split coding assistants into IDE integrations and terminal tools, and added an entirely new Images & Video category. All links and licenses were reverified; new entries (PearAI, Windsurf, Sourcegraph Cody, SuperAGI, Leon AI, Draw Things, Fooocus, StableSwarmUI, and others) were validated for active maintenance. Inclusion criteria: project is actively maintained (commits in the last 90 days), has a verifiable open-source licence or a clear commercial-use statement, and either holds meaningful user share in 2026 or fills a category that would otherwise be empty. Projects that go inactive for more than two release cycles are removed; new entrants that pass the criteria are added at the next review. To suggest a project for inclusion, open an issue or PR against the PromptQuorum repository — include the project URL, licence, and a one-sentence description in the format above.

Sources

Frequently Asked Questions

What is the difference between a local LLM runtime and a desktop app?

A runtime (Ollama, llama.cpp, vLLM) is the engine that loads model weights and serves an API — typically OpenAI-compatible. A desktop app (LM Studio, Jan, GPT4All) is a chat UI that calls a runtime under the hood. Some apps bundle their own runtime (LM Studio embeds llama.cpp), others require you to install a runtime separately (Open WebUI calls Ollama). The runtime decides what is possible; the app decides what is convenient.

Can I use multiple tools from this list at the same time?

Yes — most stacks combine 2-4 tools. A common setup: Ollama as the runtime, Open WebUI for chat, AnythingLLM for document chat, and Continue.dev for coding — all four run against the same Ollama instance on a single machine. The "Common Real-World Stacks" table above lists the recipes that work without conflict.

Which tools work fully offline with no telemetry?

Ollama, llama.cpp, vLLM, Jan, GPT4All, Open WebUI, AnythingLLM, PrivateGPT, Continue.dev, Aider, KoboldCpp, Llamafile, MLX-LM, and most of the AGPL/MIT-licensed apps in this directory work fully offline once the model is downloaded. LM Studio and several closed-source tools have optional analytics that can be disabled in settings — verify by running a packet capture once after install. Browser-based UIs (Open WebUI, LibreChat) are local-only when configured to use a local backend.

Are any of these commercial-licensed (not free for commercial use)?

A handful: LM Studio, Msty, Backyard AI, Layla, and Cursor are closed-source — generally free to use but not redistributable, and commercial terms vary. Private LLM is paid. AGPL-licensed tools (Jan, KoboldCpp, text-generation-webui, SillyTavern, Khoj, Copilot for Obsidian) are free for any use including commercial, but the AGPL terms require source disclosure if you modify and host them publicly. Apache 2.0 and MIT projects (the majority) are usable in any context including commercial without attribution constraints beyond the licence text.

Which tools support Apple Silicon (M-series chips) natively?

Ollama, llama.cpp, MLX-LM, LM Studio, Jan, Enchanted, GPT4All, MLC Chat, AnythingLLM, and most Electron/Tauri apps run natively on Apple Silicon and use the Metal backend. MLX-LM is Apple-specific and the fastest for large models on M-series. vLLM, TensorRT-LLM, and ExLlamaV2 are NVIDIA-focused and either do not run or run poorly on Apple Silicon — for Apple users, Ollama with the Metal backend is the default.

Do all these tools support GGUF model format?

GGUF is the native format for llama.cpp and any tool that wraps it (Ollama, LM Studio, Jan, GPT4All, KoboldCpp, Llamafile). vLLM and TensorRT-LLM use their own optimised formats (typically AWQ or FP16) for higher throughput. ExLlamaV2 uses EXL2 quantisation. MLX-LM uses MLX-converted weights. Most listed tools accept GGUF; a few (vLLM, TensorRT-LLM, ExLlamaV2, MLX-LM) require a one-time conversion step from the original Hugging Face weights.

Which tools are best for users with no coding experience?

GPT4All has the simplest install (one click, runs on 8 GB RAM). LM Studio is the most feature-rich without requiring a terminal. Jan is the most privacy-conscious of the no-code options. For document chat without command-line work, AnythingLLM is the easiest. All four are listed in the Desktop GUI Apps category above.

Can I run these tools on a server and access them remotely?

Most server-capable tools (Ollama, vLLM, LocalAI, Open WebUI, LibreChat, PrivateGPT, AnythingLLM) expose an HTTP API and bind to a network interface configurable in settings. Standard pattern: run Ollama on a home server or VPS, run a UI on your laptop or phone pointing at the server's IP. Treat the API like any web service — bind to localhost behind a reverse proxy, or to a private network with proper authentication. Open WebUI ships with multi-user support out of the box.

Which tools support multi-user / team setups?

Open WebUI, LibreChat, h2oGPT, AnythingLLM (with admin features enabled), and Dify are designed for multi-user use, with role-based access and per-user conversation history. vLLM is the right serving layer underneath when concurrent inference matters — it batches requests across users for throughput unattainable on Ollama at concurrency above ~3.

How often does this directory get updated?

Every six months, with mid-cycle patches in between. Mid-cycle changes (a project goes inactive, a new tool gains meaningful share, a licence changes) get patched into the existing entry. Entirely new categories (like the Images & Video category) are added during scheduled refreshes to keep the structure stable. See the "Last updated" date at the top of this page for the most recent refresh. The "Sources" section above lists the community indexes used to spot-check what the ecosystem is actually doing between refreshes.

Can I do image generation locally without cloud calls?

Yes — Stable Diffusion, ComfyUI, Invoke AI, AUTOMATIC1111 WebUI, and others in the Images & Video category run entirely on local hardware. Stable Diffusion needs 6+ GB VRAM (RTX 3060, RTX 4060, or equivalent); Fooocus and other optimized UIs can run on cards with 4-6 GB. Real-ESRGAN upscales generated images; ControlNet adds spatial control (edges, poses, depth maps); AnimateDiff generates video from text. All run without sending data to external servers.

Browse the slides below or download as PDF for offline reference. Download Reference Card (PDF)

About this data

This directory is compiled with AI-assisted research from public sources (project READMEs, official sites, GitHub repositories). It is not exhaustive and may contain errors — hardware requirements, pricing, and platform support change frequently and some fields have not been independently verified yet.

Most entries carry no status badge: they are listed from public sources but not independently reviewed. A "Verified" badge means core facts (license, platforms, pricing) have been manually checked. A "PromptQuorum-tested" badge is only shown for tools PromptQuorum has actually installed and run — most entries do not have this badge yet.

The "PromptQuorum-tested" badge is reserved for tools we have personally installed and run — an untested tool never carries it, regardless of popularity or stars.

Spotted something wrong? Email hello@promptquorum.com and we will correct it.

← Back to Power Local LLM