Get access to models from OpenAI, Anthropic, Google, and others — all in one place. Automatically pick the best models for each task, see them side-by-side to compare, and merge them into a comprehensive answer.
Subscribe
Sign in
Get access to models from OpenAI, Anthropic, Google, and others — all in one place. Automatically pick the best models for each task, see them side-by-side to compare, and merge them into a comprehensive answer.
ChatBetter
Hey Product Hunt! I'm Noah, co-founder of Unsupervised and one of the makers behind ChatBetter.
We think people should be using more LLMs: If you want to get the best results out of AI, you can't use just ChatGPT. We found that using the right model for the task at hand had a pretty big impact on output — models have more differences than most people think and they each excel at different tasks. We made ChatBetter initially for internal use for this reason, but are launching it to the world because we realized how needed it was!
We're building ChatBetter to make it really easy to use more models than what ChatGPT gives you (and get more right answers because of it).
Here's how:
We give you access to all of the major AI models (Claude, Perplexity, Gemini, etc) from a single interface
ChatBetter predicts the best models and the right reasoning level to use on every request
Compare the top 2-3 responses side-by-side from different makers on every query
Merge the responses into a comprehensive answer
It does more as well : you can organize chats by project/client with folders, use searchable history, manage one context window across different models, use memories, edit the information in the context windows during the conversation, all sorts of stuff.
For Teams: ChatBetter is ready out of the box with features including custom branding, SSO integration (Okta, Rippling, Microsoft), company context personalization, and robust privacy and security controls (MFA, IP restrictions, audit logs) + more!
If you're willing to give it a try, I'd love to hear what you were able to do with it! noah@unsupervised.com.
Try ChatBetter for free! www.chatbetter.com
You can choose between free and Pro ($20/month) plans.
Raycast
@noah_horton appreciate the value of getting "multiple opinions" from the different LLMs! I'm often left wondering which I should use, and occasionally will run the same prompt multiple times. This is so much easier!
Do you think this comparison is a good idea for all prompts — or only certain high value or complex prompts?
ChatBetter
@chrismessina thanks Chris!
We have found that even simple prompts often benefit. You sometimes get similar answers from all the models, but even then it is nice confirmation. But even for simple queries, we have been surprised by how often the answers differ.
What's most important about the above is how it affects AI adoption; we see that a wrong answer on a question from chatGPT would cause many people to stop using it for some period. But when you have multiple answers, those users almost never get stopped!
Which model of Openai will I receive with this subscription? I'm comparing the cost of a Plus subscription to that of a Chatbetter subscription.
ChatBetter
@michael_vavilov we allow access to all their models. That includes premium ones like o1! Same for the other labs.
ChatBetter
@michael_vavilov All the major ones (GPT-4.5, GPT-4.1, o3/o4, etc.) — and we add new models as they add them to the API :)
How can users efficiently manage conflicting answers from different AI models in ChatBetter to avoid time-consuming evaluations while maintaining output reliability?
ChatBetter
@vouchy no evals needed! We return multiple responses to the user in an easy way where they can scan them in seconds. We discovered that people love seeing the options. If they are similar, you see it instantly, and if they differ, you can choose what you like.
ChatBetter
@vouchy My two favorite things are that ChatBetter:
Shows you multiple responses side-by-side and
can merge answers from multiple LLMs (which can highlight agreements or disagreements between models).
There are a lot of small features you can use to change results, quickly pick responses to focus on, etc. — but those kind of fade into the background until you want to use them.
In practice, this is something you get used to super fast when you use it and then can't imagine going back.