SpeakGPT

Teslasoft

Contains ads

About this app

SpeakGPT is a flexible, OpenAI-compatible client for working with AI APIs. It combines a chatbot and voice assistant in one app, making it easy to write, explore ideas, or learn something new hands-free or through chat. The app is designed to work with a wide range of models - including GPT, LLaMA, Mixtral, Gemma, Gemini, vision models, DALL·E, and more - while also letting you connect custom endpoints and experiment with different providers.

Features list:
● Privacy focused
● Chat (chat history is saved locally and can be imported/exported is needed)
● Images generation (DALL-e* or GPT Image 1**)
● Images recognition (use your photos and images with AI)
● Custom AI models (fine-tuned) are supported
● Activation prompt
● System prompt
● Assistant
● SpeakGPT in context menu
● SpeakGPT in Share sheet
● Prompts Library
● A lot of different models (including latest flagship models like gpt-5, o1, o3, o4, gpt-4.1 and gpt-4.5***)
● No annoying captchas or paywalls
● Pay-as-you-go system
● Tips for newbies
● Advanced support
● Different API providers
● Customize AI model parameters****
● Compatible with OpenRouter

This app is open-source. Report bugs, ask a questions and view code here: https://github.com/AndraxDev/speak-gpt

* DALL-E models will shutdown at May 12, 2026. Until that date, you can still access that models from SpeakGPT.

** GPT Image 1 is a flagship model that may not be available for all users. You may need to verify your organization at OpenAI to use this model.

*** Some of these models may require you to have higher OpenAI usage tier.

**** Please refer documentation for AI model you are using to learn which parameters are supported.

DISCLAIMER:
This app uses API access. Some API provides use pay-as-you-go subscription. You will need to provide your API key to use this app. Some models may not be available with this app or/and may require additional identity verification for their providers.

If you chose OpenAI as your API provider:
You can revoke your OpenAI API key anytime at: https://platform.openai.com/api-keys

Learn about pricing here: https://openai.com/pricing

For more providers please refer it's documentation, Terms of Service and Privacy Policy.


Information for users who want to use Google Gemini API with this app:
SpeakGPT does not officially support Google API keys, but you can still use Google Gemini using OpenRouter API. View more info at: https://openrouter.ai/docs#models

Ratings and reviews

I’ve been using Speak GPT for a while now, and honestly, it’s been great. What I really like is that it lets me use my own API key, which gives me full control. The app is smooth, the responses are quick and accurate, and the interface is clean and easy to use. I use it every day—especially for studying, or just casual chatting. It’s reliable and does exactly what I need. Definitely recommend it for using api in mobile. Big thanks to the developer for making such a useful app!

I'm enjoying! I would like to request the addition of a voice activation feature to make it even more convenient and user-friendly. Keep up the great work!

Smooth app with great features. Definitely one of the best open source ChatGPT clients. Using as assistant as well!

What’s new

- Fixed bug when API endpoint was not updated properly in the chat settings
- Added GPT Image 1 Mini model
- Set GPT Image 1 mode as default model for image creation
- Updated dependencies
- Other minor bugs fixed

Important note: GPT Image models may not be available for all users. Availability may depend on region, usage tier or OpenAI account verification status.