Last updated: 9/5/2024
Visit https://livebench.ai to view the top ranked models.
Table of contents
Intro with AI
If you have used ChatGPT and got amazed how it changed your life forever, and its launched have hyped the industries rapidly integrating and improving with AI in various ways.
Exploring ChatGPT alternatives
If you have used ChatGPT, you may have find that some features like Image and Document Analysis, Web Browsing, GPTs can be limiting and falls you back with the dumber model. But have you thought of other alternatives?
Many of these tools are free to use, with limitations, some paid offerings are mentioned as part of a limitation. Here are the Good ChatGPT alternatives and other AI tools:
Chat
Google Gemini
Link: https://gemini.google.com
Lets you access to Google’s in house LLM models (currently Gemini). Has internet access, extensions, and native multimodality which can accept images and voice as input, and image generation capabilities.
Tips:
Enable extensions to work with your email, files, navigation or youtube
- What’s hot in my inbox
- Can you find me a document that relates to research (this includes files who shared with you)
- Find me a YouTube video about retro computing
- Summarize this video: https://www.youtube.com/watch?v=nUFgBhBBB-E
- Create an image of fish
Upload images to let Gemini understand your images that you cannot describe.
When you subscribe to Gemini Advanced, you also get additional features such as:
- Access to Gemini 1.5 Pro with long context (1M tokens)
- Document Analysis
- Gems (Gemini with specific skills and personality)
- Gemini Live
Google AI studio
Link: https://aistudio.google.com
You can also use Google AI studio to access latest SOTA Gemini models for free but with limitations (rate limits) if you cannot afford Gemini Advanced, and experiment with capabilities.
If you’re an AI developer, this is the best choice!
Claude
Link: https://claude.ai
In 2024, Claude releases Claude 3 model family, along with the latest 3.5 sonnet which is available and the current model used in free tier. Claude 3.5 sonnet beats all models including GPT-4, Gemini, and others… Excels at complex tasks like math and coding and others like logic, reasoning, common sense.
It also has document and image analysis which is native to the family of Claude 3 models.
Limitations:
- Extremely low on usage caps for free tier unless you pay
- No internet access
Le Chat Mistral
Link: https://chat.mistral.ai
Lets you access models from Mistral serving as ChatGPT alternative
Offers models for free include:
- Large 2 - aims to be on par with GPT-4o
- Codestral - aims to specialize at coding
- Nemo - meant to be fast and cost effective
Limitations:
- No internet access
HuggingChat
Link: https://huggingface.co/chat/
Enables you to access free and open-source LLMs (or Large language models) as an alternative to ChatGPTs or Google’s Gemini with variety of models optimized for specific fields.
Some models like LLAMA 3.1 can be on-par with ChatGPT 4o model. It can also generate images, browse the web, parse documents to the model, edit images using words, and more depending on models selected.
You can try it for free as guest or enable history by signing up for free
Creativity
Tools that goes beyond text generation!
Microsoft Designer
Link: https://designer.microsoft.com
Create social media posters, designs, banners, or edit photos with AI.
Notable features:
- Image Creator - using the latest DALL-E 3 model
- Design Creator - create and describe your design with just your words
- Generative Erase - erase unwanted objects in your image
- Sticker Creator - create stickers using the latest DALL-E 3 model
- Album creator
You can also manually design your creations after you asked AI to make it or start from scratch.
Limitations: Microsoft Account Required for AI features.
ImageFX
Link: https://aitestkitchen.withgoogle.com/tools/image-fx
Access to latest Google’s Imagen models, as of this writing, it serves Imagen 3 model which produces realistic and high quality images, aiming to compete with DALL-E 3 and Midjourney.
https://deepmind.google/technologies/imagen-3/
Limitations:
- Has high moderation that not all images can be created.
- Usage limits per day.
Suno AI
Link: https://suno.ai
Generate music, complete with voiced lyrics. Just ask AI what music you want to make based on your thoughts, or manually dump a lyrics that you composed on your own and ask AI to refine it or create it, no need to have musical background, just tell AI what and how you feel, what music you want to make, and more.
Limitations: You can create 10 (100+ in some cases) songs per day on free tier
MusicFX By Google
Link: https://aitestkitchen.withgoogle.com/tools/music-fx
Generate music with the variety of instruments
Limitations: It doesn’t produce vocals compared to suno and is limited upto 1:30 music length but DJ mode can stream music upto an hour
Gemini in Google Docs, Slides, Sheets, Meet, Vids using Workspace Labs
Link: https://support.google.com/docs/answer/13447104?visit_id=638424634089901740-2280639011
Available in Limited Preview
Enables you to draft, create, and organize using your favorite Google Workspace apps.
It unlocks the Gemini chat sidebar which can be pressed with sparkle symbol
The sidebar chat experience can answer questions about your files, create images, summarize emails, and more. It is accessible on Web and across Google workspace apps including Google Drive.
Currently you can try using Labs features using these apps on the desktop
- Google Docs - Help me write (drafts paragraph)
- Google Sheets - Help me organize (drafts tables)
- Google Slides - Create images using Imagen 2 model
- Google Meet - Create background images using Imagen
- Unlocks Google Vids - AI powered collaborative video maker with image generation feature (https://vids.google.com)
For Android/iOS
- Google Keep - Help me create lists (Android app)
- Gmail - Help me write and summarize email button with chat (chat experience can also be activated using the Sparkle icon).
Limitations: See limitations above in the link!
Elevenlabs TTS
Link: https://elevenlabs.io
Generate natural sounding text to speech voices with collection of voices library, text-to-speech and speech-to-speech (voice tone replication with voices library).
Limitations: Voice cloning requires paid tier, and you can generate voices upto 10000 character limit per month.
More AI
Google AI Labs
Link: https://labs.google
Limitations: Some AI tools might require you to use such as US VPN, Google Pixel device with AI capabilities (notably the Pixel 8), and Google one subscription. This can subject to change.