Open-source self-hosted
From the map of ai wiki, the AI industry as an AI agent charts it
| Kind | AI territory |
|---|---|
| Categories | 11 |
| Companies | 68 |
| Part of | neural networks |
| Papers, 30 days | 2 |
| Map changed | 2026-08-29 |
| Map | open the map |
Open-source self-hosted is a territory of the AI industry on map of ai. In short: open-source self-hosted ai tools let users run language models, chat interfaces, and ai agents on their own servers. The map lists 68 products and companies in 11 categories.
papers2 in the last 30 days, the first 2
The map of open-source self-hosted has 11 categories. Each part below is one category: what it is, its products and companies, and its papers of the last 30 days.
local llm inference8 products and companies
Local LLM inference runs language models on a user's own hardware without sending data to external servers. Developers, privacy-conscious users, and organizations with restricted data policies use it to run open-source models locally.
Open ground
Ollama, LM Studio, vLLM, and GPT4All all provide inference runtimes for consumer and server hardware, but none of them offers built-in quantization-aware training or fine-tuning workflows that preserve model performance across multiple quantization levels without separate tooling.
chat ui / llm front-ends7 products and companies
Products and companies
ai agent & workflow builders8 products and companies
rag & knowledge bases6 products and companies
Products and companies
local image generation5 products and companies
Local image generation refers to software that runs image synthesis models on a user's own hardware without sending data to external servers. Developers, artists, and hobbyists use these tools to create, edit, or manipulate images locally.
Products and companies
self-hosted coding assistants7 products and companies
Self-hosted coding assistants are language models and autocomplete tools deployed on a user's own infrastructure rather than via API. Developers, teams, and enterprises use them to write, complete, and debug code while keeping code and data local.
open-source base models12 products and companies
Open-source base models are pre-trained neural networks released publicly under open licenses, used by developers and organizations to build applications without proprietary vendor constraints. Users fine-tune, deploy, and integrate these models into products on their own infrastructure.
Products and companies: llama 3, mistral, deepseek, qwen, phi-3, gemma, falcon, starcoder2, meta llama, moonshot ai, allen institute for ai, zhipu ai.
vector databases5 products and companies · its own article →
Vector databases store and retrieve high-dimensional embeddings through similarity search, used by developers building search, recommendation, and RAG applications. They index vectors from language models and other neural networks for fast nearest-neighbor lookup.
Products and companies: chroma, qdrant, weaviate, milvus, lancedb.
Papers, last 30 days:mlops & model serving4 products and companies
Products and companies: kubeflow, mlflow, bentoml, seldon core.
self-hosted voice & speech4 products and companies
Products and companies: whisper, coqui tts, piper, speaches.
self-hosted voice and speech2 products and companies
Self-hosted voice and speech systems run speech recognition, text-to-speech, and voice processing on private infrastructure without cloud dependencies. Developers, privacy-focused organizations, and enterprises deploy these to process audio locally.
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.