More features with Geeps Plus
A one-time upgrade that improves your experience with additional features and supports future development. No subscription required and applies to all devices.
Become a supporter with an in-app purchase eligible for Family Sharing.
Black Friday Sale: Get Geeps Plus for only $9.99!
Multiple model providers
Geeps Plus users have built-in support for OpenAI, Anthropic, Gemini API and OpenRouter plus multiple custom providers:
- OpenAI via Chat Completions API. Support for Responses API will be added in a future update.
- Ahthropic Claude API to chat with Claude Sonnet, Haiku and Opus models
- Gemini API - using OpenAI compatibility mode.
- OpenRouter API to chat with hundreds of models from OpenAI, Anthropic, Google, Mistral, DeepSeek, Perplexity, Meta, Cohere and others.
- many OpenRouter models are free to use!
- NEW! Add multiple custom providers that are synced across devices with iCloud.
Custom providers must support OpenAI compatible Chat Completions endpoints. Most of LLM providers do, including xAI (Grok), Groq, DeepSeek, MS Azure, Mistral, Cohere, Perplexity, Moonshot AI, TogetherAI and many others.
You can easily switch between providers by long pressing on the compose button (right click on macOS).
Chat with PDFs and other files
Attach text based files to context and let the model respond based on their content.
This feature works by extracting text from the attached files and adding it directly to the context window. Be mindful of the context window size. If your files contain images, those can not be extracted but you can attach them separately in the same chat if you need.
- Supported file types include: PDF, TXT, DOCX, RTF, HTML, Markdown and some others where text can be extracted (e.g. JS, JSON, CSV, CSS, etc.)
- Attach files from Files/Finder or drag and drop into the chat on macOS
- The files are not stored anywhere in the app and are not uploaded to the API. They are only used to extract text on device and add it to the chat. When chat is deleted the extracted text is deleted as well
Using models with vision
Most OpenAI, Anthropic and OpenRouter models support vision.
Use models with vision for tasks like:
- subject recognition (animals, plants, objects, geographical locations, etc.)
- image captioning, description and analysis
- translation of text in images
- image classification
- prototype design from sketches
- and other image-related tasks
Some reasoning models may not support image inputs and will return an error. Images are resized before sending to the API and are seamlessly synced with iCloud.
Image generation with GPT Image and DALL-E
Image generation currently works with OpenAI GPT Image, DALL-E 2 and DALL-E 3 models.
- GPT Image can be used in Create or Edit mode to generate images based on text prompts or to edit existing images.
- you can start in Create mode to generate an image and then switch to Edit mode to use the generated image as input and continue refining it. In Edit mode you can add image attachments to combine multiple images into one or use them as elements in the new image.
- you can also start in Edit mode by selecting an image(s) from your library or Files/Finder and then use it as a base for further edits.
- GPT Image supports square, landscape and portrait aspect ratios with up to 1536x1024px resolution.
- with GPT Image you can control transparency, background* and moderation settings.
- Dall-e 2 supports generating up to 10 images at a time with 1 to 1 aspect ratio and up to 1024x1024 resolution.
- Dalle-3 supports generating one image at a time with 1x1, 9x16 or 16x9 aspect ratios and up to 1792x1024 resolution great for creating wallpapers.
- Dalle-3 also supports vivid or natural style and standard or hd quality.
Pricing
Billing for these features is based on your token usage and is managed through your provider dashboard.
Geeps Plus is a one-time in-app purchase that costs $14.99 $9.99! (in the US). Prices vary a little depending on your country and local taxes. Check more details on the pricing page.