A new large language model trained with reinforcement learning to perform complex reasoning. o1 thinks before it answers—it can produce a long internal chain of thought before responding to the user.Here it is — Q*. AI that reasons... the singularity grows closer. 🍓@chrismessina saw the videos on Twitter yesterday! The pace of AI is insane, you never know what will be possible just round the corner@chrismessina Complex reasoning made simple with O1—AI just got a whole lot smarter! ✨🔍 Excited to see where this leads!@chrismessina 😸 We won't know when singularity hits, because it will be named ChatGPT_core _LLM_4o16 large-model_preview_1064 in the drop down menu, mixed together with the 200 other models we can choose from 😅The real question is - how long it takes to solve a hard problem and how many tokens it would take...Can I ask the difference between o1 and claude-3.50-sonnet? A step function difference or just slightly better?Upvoted, big fan of OpenAI. But not so sure that this time o1 can actually beats Claude 3.5-Sonnet. What would be your thoughts dear hunters? What is your best use cases for OpenAI o1?

OpenAI o1

Boost your customer support with CX Genie. The AI-driven platform integrates chatbots, ticket management, help desk, and more to automate tasks and personalize customer interactions. Reduce support costs, improve efficiency, and grow sales for your business.

CX Genie

Automation Workflow Builder creates AI agent teams with access to LLM models and integrates with Zendesk, Intercom, WhatsApp, and websites. It offers features like embed options, link copying, hallucination prevention, and LLM cost optimization.Hi, nice to meet you!
I'm Roy Nativ, Founder of Officely AI.

At Officely AI, we transform any process into an AI-driven operation with our innovative Team AI Builder. Unlike traditional models that rely on a single AI agent, our platform utilizes a team of specialized AI agents—each with unique personalities, objectives, and permissions, using models from GPT to Claude and LLAMA available on Hugging Face. This setup enhances problem-solving accuracy and reliability.

Our AI agents also communicate dynamically, sharing insights to deliver the most effective responses quickly. This collaborative approach is crucial for providing finely-tuned, contextually appropriate solutions.

I’m here and eager to discuss how Officely AI can transform your business operations and to answer any questions you might have about getting started!@kshitij_mishra4 thanks! Appreciate 🙏🏻Congrats on the launch 🎉! You’ve created something truly special here. I’ll definitely use it and look forward to seeing its success! 🙌🔥 @roy_nativ@zulkarnaim thanks! Appreciate 💚🙏🏻I came across this through an ad and gotta say, looks pretty cool but I’m wondering, how does the team AI thing actually work in real time, like if two agents disagree on a solution what happens then, andalso, will it make it more expensive to use multiple models instead of one, just curious about the practical side of it instead of all the big promises@codecrafteddev I’m glad you asked! A language model has many issues in a business context (for example, hallucinations, max token limitations). The right way to manage this is to split the task among several different agents. One agent based on GPT formulates search queries for RAG, another agent based on Claude crafts the response, and then another agent, based on a different model, verifies the answer and assigns a score.

Each agent has its own role. The customer feels like they’re interacting with a single agent, but behind the scenes, there’s an entire team at work. This approach also helps reduce costs by allowing the use of a cheaper model in certain cases@codecrafteddev https://youtu.be/573AFURU4EU?si=...Hey Roy,
curious how your AI teams will really handle complex workflows or just another buzzword filled tech piece, what about real-world integrations, do these agents really sync seamlessly with platforms like Zendesk and Intercom or are users gonna face endless bugs and lag, hope to see some solid proof soon@contentguruny Hey Mark,

Great question! Our AI teams are built to handle complex workflows effectively, not just buzzwords. We’ve integrated seamlessly with platforms like Zendesk and Intercom.

In Zendesk, for example, we automatically scan the help center, and users can trigger our AI with an “Answer This Ticket for Me” button. The AI formulates a response, and we then track how users interact with it—whether they leave it unchanged, tweak the tone, or rewrite it completely. For cases where the response remains unchanged, we convert them into bot-driven solutions, ensuring efficiency and relevance.

Stay tuned for some solid proof—we’ve got exciting case studies already published !

Best,
Roy

Officely AI

Anyone can create an AI character based on their interests, and creators can build an AI extension of themselves. Start chatting with these AIs in Instagram, Messenger, and WhatsApp. *only available in USA for nowThis is really cool! I am interested in seeing how people interact with AI on social media and how natural the AI sounds like.Hi, I am from India and i am interested in this product, this there a way you see to release it to other country in your roadmap.How do you ensure the AI characters remain distinct from real users, especially for creator-based AIs? Addressing those concerns will be crucial for user trust.Meta AI Studio is a cool way to bring your thoughts to life through AI characters. The idea of creating personalized AIs to chat with on popular platforms like Instagram, Messenger, and WhatsApp is exciting. Perfect for those looking to add a unique twist to their conversations! Can't wait to see it grow!

Meta AI Studio

Not Diamond isn’t like other chatbots you’ve used. Not Diamond automatically calls the best model for any prompt and improves in real-time based on your feedback, continuously learning your preferences. Not Diamond is the last chatbot you’ll ever need.Hey Product Hunt!

I’m Tomás, co-founder and CEO of Not Diamond: chat.notdiamond.ai

Not Diamond is the first AI chatbot that automatically recommends the best AI model for every message and improves in real-time based on your feedback, continuously learning your preferences for when to use which model.

The future is multi-model. Every week (often every day), a new state of the art AI model is released. These models vary widely in their abilities across domains, and over the past year I found myself constantly jumping back and forth between chatbots, manually comparing different models to get the best answer. More complicated still, my preferences kept evolving over time as the models themselves did.

We built Not Diamond because we want to take the guesswork out of AI. Now you can leverage the best AI models with the keen-eyed discrimination of a veteran AI engineer. Even better, Not Diamond learns your preferences for which model to use on everything from conversation to coding to creative writing. 

This is an incredibly powerful tool packed with features—here are a few of my favorites:

🍢 Automatic AI model routing: Not Diamond automatically chooses the best LLM for every message you send, with new models added every week.

💫 Real-time learning: Unlike traditional chatbots, Not Diamond learns from your feedback in real-time, ensuring that your experience is always getting better.

⚔️ Arena mode: Turn on arena mode to compare models head-to-head in a blind selection process, supercharging your personalized model recommendations.

🔥 SOTA quality: Not Diamond beats every individual LLM on every major benchmark and continuously improves as new models are released.

🛹 API access: Not Diamond is also available via our API so that you can implement automatic AI model routing in your own AI applications.

🪩 Open-source: Our chat app is based on an incredible open-source template built with the Not Diamond API by McKay Wrigley.

🌸 Free to use: Not Diamond is entirely free to use, with no limits on how many messages you send to our chat app.

I’m super excited to share this with you all—I’ve been using it nonstop and it’s a game changer. Try it out and lmk what you think!

chat.notdiamond.ai@tomashernandokofman Congrats on the launch, very cool.  I use Abacus.ai which allows you to choose between different LLM model, but I like that this gives you the best model based off the task at hand, takes the choice out of my hand, I don't always know what the best option is, but also like that I can thumbs down it to express my displeasure with the result!  Very well done! Wish you all the best! ¡Bien hecho!@mogabr muchas gracias! I'd love to make manual model select dropdowns a thing of the past : )Looks promising trying to orchestrate across models. What’s the logic though? Under what conditions does the model switch ?@kaushik_mukherjee1 Great question. We base model recommendations on a rigorous, large, cross-domain dataset that evaluates different LLMs based on message semantics, domain, intent, length, and complexity—our team had been publishing in top AI journals for over 10 years, so this is the culmination of a lot of intense research effort!@kaushik_mukherjee1 @tomashernandokofman That's impressive! Can you share more about the specific criteria or metrics you use to evaluate the models? I'm curious about how you ensure the rankings remain fair and unbiased across such a diverse dataset.@kaushik_mukherjee1 @divy_divya_c we haven't released a technical report yet, but you can see that Not Diamond outperforms every single LLM on every single major benchmark: https://notdiamond.readme.io/v1....

Importantly, while we provide a really strong solid baseline for model recommendations when you first sign up, we don't assume there's a "one-size-fits-all" set of recommendations here, which is why we hyperpersonalize routing to your preferences based on the feedback you provide.@kaushik_mukherjee1 Cheers to the team for launching this! Wishing you an extraordinary journey ahead. What’s the story behind the product's name?@kaushik_mukherjee1 @divy_divya_c @tomashernandokofman impressive results. Technically speaking you are an intelligent LLM router, right ?

I am curious to understand the biases and weaknesses of each frontier model. As an AI enthusiast, I would definitely read a newsletter with key insights, if you have one.Congrats on the launch. Don't you think that LLM choices will be irrelevant in the future? As is, for example, technologies behind POS payments, etc. Also do you target b2c or b2b with this app?Thanks @sasha_briu ! With new LLMs released every week, including models specialized for specific domains, I believe machine intelligence will become increasingly diversified over time. The same pattern of specialization has played out in human intelligence and biological systems, and generally the universe tends towards it as a highly efficient pattern for distributing capability. Most foundation models today are actually already routing under the hood using Mixture of Experts architectures, which routes to multiple expert models. 

Our chatbot is for consumers, but businesses can also leverage the API behind the app to provide automatic model routing within their applications.I've following Not Diamond for a while and first of all congrats to the team to bring this product to live!!

I'm very excited to try it out for our product at Opencord AI! Love the concept of AI model router, and this is exactly what all AI applications needed. 

No need to evaluate new LLMs is a HUGE relieve from thousands of AI apps founders/builders (coz our time prob better spend on finding the GTM or PMF)@kellyann3644 thank you—so excited to share this with you!

Not Diamond

Automate 80% of Customer Inquiries in 80+ Languages with Kotae – Your AI Chatbot Powered by Your Content. Ready in 5 Minutes, No Coding Required. Enjoy a Free 30-Day Trial with Personalized Onboarding.

Kotae

GPT-4o mini scores 82% on MMLU and currently outperforms GPT-4 on chat preferences. It is priced at 15¢/million input tokens and 60¢/million output tokens, an order of magnitude more affordable than previous frontier models and 60+% cheaper than GPT-3.5 Turbo.

GPT-4o mini

The lowest latency conversational AI ever released.Moshi can perform small talk, explain various concepts, engage in roleplay in many emotions and speaking styles.

Moshi

Time is money. Why waste it on tedious and repetitive tasks? With MindPal, in just 5 minutes, you can build custom AI tools that automate anything, freeing your team to focus on what truly matters. A fast track to superior efficiency.

MindPal

Krater.ai's "Chat with AI" app now lets you chat with 15+ AI models from 8+ AI research companies, including GPT-4o, Claude 3, Gemini 1.5, Perplexity Sonar + More. All inside one interface, starting at $7/mo.