PrivateGPT

AI & LLMs

Interact with your documents using the power of GPT, 100% privately, no data leaks

Latest: v1.0.15 releases2 breaking changes1 common errorsUpdated Jun 18, 2026View on GitHub

Release History

v1.0.14 fixes

Jun 18, 2026

Version 1.0.1 addresses several bugs, including missing Docker dependencies and optional embedding API keys, and standardizes local data paths using PGPT_HOME. A performance improvement was also made to prompt caching using date-level current_date.

v1.0.0Breaking10 features

Jun 3, 2026

PrivateGPT 1.0 merges two years of development, transforming the project from a semantic search pipeline into a full application backend that connects to external inference servers for production AI applications.

v1.0.0-rc1Breaking4 fixes6 features

Jun 2, 2026

This release introduces a major revamp to PrivateGPT (v1.0.0-rc1), adding MistralAI support, Ollama connection retries, and dependency updates, alongside several bug fixes.

Common Errors

ResponseHandlingException1 report

ResponseHandlingException in privateGPT often arises from excessively long processing times during document ingestion or query processing, exceeding default timeout limits. Resolve this by increasing the timeout values in your configuration files (e.g., `settings.yaml`) for relevant modules like the embedding or LLM components. Alternatively, optimize your document processing pipeline to improve efficiency and reduce processing time for large or complex files.

Related AI & LLMs Packages

AutoGPT

AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.

Ollama

Get up and running with OpenAI gpt-oss, DeepSeek-R1, Gemma 3 and other models.

LangChain

🦜🔗 The platform for reliable agents.

ComfyUI

The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface.

llama.cpp

LLM inference in C/C++

GPT4All

GPT4All: Run Local LLMs on Any Device. Open-source and available for commercial use.