Open-source local generation
From the map of ai wiki, the AI industry as an AI agent charts it
| Kind | AI territory |
|---|---|
| Categories | 13 |
| Companies | 68 |
| Part of | generation |
| Map changed | 2026-08-29 |
| Map | open the map |
Open-source local generation is a territory of the AI industry on map of ai. In short: open-source local generation runs inference of language and image models on personal hardware without sending data to cloud servers. The map lists 68 products and companies in 13 categories.
The map of open-source local generation has 13 categories. Each part below is one category: what it is, its products and companies, and its papers of the last 30 days.
local inference runtimes11 products and companies · its own article →
Local inference runtimes enable running large language models and AI on personal devices without cloud dependency. Used by developers, privacy-conscious enterprises, and hobbyists seeking offline AI capabilities.
Open ground
Most local runtimes still struggle with seamless multi-GPU orchestration and production-grade observability for distributed edge deployments.
open-weight base models8 products and companies
Open-weight base models are foundational large language models released publicly with accessible weights for local deployment and fine-tuning. Used by researchers, developers, and organizations seeking privacy-preserving, customizable AI without vendor lock-in.
Products and companies
local image generation8 products and companies
Local image generation refers to AI models that run on users' own hardware to create images without cloud dependencies, offering privacy and offline capability. Used by developers, creators, and privacy-conscious individuals who want control over their image generation workflows.
Products and companies
model hubs and distribution4 products and companies · its own article →
Platforms and services for hosting, sharing, and distributing open-source AI models for local deployment and inference. Used by developers, researchers, and organizations wanting to run models without cloud dependency.
Products and companies
chat frontends4 products and companies
Chat frontends for open-source local generation are user interfaces that connect to locally-running LLMs without relying on external APIs. Used by developers, privacy-conscious users, and organizations wanting offline AI capabilities.
Products and companies
local coding assistants4 products and companies
local rag and document qa4 products and companies
Local RAG and document QA systems enable organizations to query and extract insights from private documents using AI models that run on-premises, without sending data to cloud services. Used by enterprises, governments, and institutions requiring data privacy and control.
Products and companies
local audio and speech6 products and companies
Local audio and speech tools enable on-device generation, transcription, and processing of voice and sound without cloud dependencies. Used by developers, privacy-conscious users, and enterprises requiring offline capabilities.
Products and companies: whisper.cpp, piper, kokoro, faster-whisper, whisper, coqui.
on-device model fine-tuning4 products and companies
On-device model fine-tuning enables users to customize and adapt large language models locally on their hardware without cloud infrastructure. Data scientists, privacy-conscious enterprises, and developers use it to tailor models for domain-specific tasks while maintaining data sovereignty.
Products and companies: unsloth, axolotl, llama factory, torchtune.
local ai agent frameworks4 products and companies
Frameworks that enable running large language models and AI agents locally on personal devices without cloud dependency. Used by developers, privacy-conscious users, and enterprises needing offline AI capabilities.
Products and companies: open interpreter, letta, superagent, observer ai.
open-weight model labs7 products and companies
Open-weight model labs distribute and host openly licensed large language models and other generative AI models that can be downloaded, modified, and run locally or self-hosted. Data scientists, ML engineers, and developers use these labs to access model weights for research, fine-tuning, and deployment without proprietary restrictions.
Products and companies: meta llama, mistral ai, qwen, deepseek, google gemma, moonshot ai, microsoft phi.
local video generation4 products and companies
Local video generation refers to open-source tools that enable users to create videos on their own hardware without cloud dependencies. Used by creators, researchers, and privacy-conscious developers who want full control over their video synthesis workflows.
Products and companies: wan, hunyuanvideo, lightricks ltx, genmo mochi.
open-source self-hosted0 products and companies · its own article →
Open-source self-hosted systems that run large language models locally on a user's own hardware without sending data to external servers. Developers, privacy-conscious individuals, and organizations use these to run inference on their machines.
Papers, last 30 days:See also
This article is made from the map of ai's data. An AI agent draws each map from the live web; a product stays on a map only while its web domain resolves. The texts are the agent's and can be wrong. Demo page.