PrivateGPT: Offline AI Chat3.3 - App Store
You'll be able to see results in few seconds.
Official App Store Link
We do not host PrivateGPT: Offline AI Chat on our servers. We did not scan it for viruses, adware, spyware or other type of malware. This app is hosted by Apple and passed their terms and conditions to be listed there. We recommend caution when installing it.
The App Store link for PrivateGPT: Offline AI Chat is provided to you by soft112.com without any warranties, representations or guarantees of any kind, so access it at your own risk.
If you have questions regarding this particular app contact the publisher directly. For questions about the functionalities of soft112.com contact us.
Publisher Description
PrivateGPT is a private, secure, and powerful AI Assistant. The app runs AI models locally on your device, so you can chat with AI even without Internet connection.
Features:
. Support the latest AI models: DeepSeek, Llama, Gemma, Qwen, etc.
. Import your own GGUF models downloaded from Hugging Face.
. For each model, choose the quantized version that works well on your device.
. Chat History saved locally.
Models:
. DeepSeek-R1
. Llama 3.1
. Qwen 3 and 2.5
. Gemma 3
. Gemma 3n E4B
. Mistral 7B
. Dolphin 2.6 Mistral 7B
. Nous Hermes 2 Mistral 7B
. SmolLM 135M
. LFM2-VL 1.6B
. IBM Granite 4
. (You can also import your own GGUF models to the app)
Privacy:
. No message is ever sent to a server. Your chats are completely private.
. No internet required. App works fine in Airplane mode.
. Zero identity tracking - you stay anonymous.
Disclaimers:
. PrivateGPT works the best on iPhone 12 or newer models, because it requires more than 4GB RAM. Older devices may not be able to run the app smoothly.
. All users can use all features of the app. To remove ads/paywalls, please subscribe to a plan.
Contact us --
Any questions about Reacts? Please contact us at team@morningavocado.com
Terms of Use --
https://www.apple.com/legal/internet-services/itunes/dev/stdeula/
Version History You can now import your own GGUF models! 3.3 10/08/2025 You can now choose from more AI models! 3.2 10/03/2025 Significantly improved performance, bug fixes 3.1 09/07/2025 Create your own AI characters! 3.0 08/20/2025 DeepSeek-R1 is now supported! Run DeepSeek-R1 Distill Qwen 7B on your iPhone today 2.3 02/16/2025 We are continuing to improve the app! New in this version: bundle a default LLM (Dolphin 2.6 Mistral 7B) into the app 2.2 05/13/2024 We are continuing to improve the app! New in this version: support more LLM models: Phi 3 mini, Llama 3, and more! 2.1 05/11/2024 We are continuing to improve the app! New in this version: support multiple models, customizable prompt templates 2.0 05/07/2024 We are continuing to improve the app. New in this version: bug fixes and performance optimization 1.3 01/30/2024 1.0 01/23/2024
