Johan Eliasson’s Post

🦾 GPT-4o Outperforms GPT-4T and others Here ⬇ you can see the full details of the Model evaluations as well as its capabilities. https://lnkd.in/eZugWRAB

How are you using AI in your business?

Mixture of Agents (MoA) combines multiple AI models to improve their overall performance. It uses several layers where each layer refines the output of the previous one, taking advantage of each model's strengths. This method leads to better accuracy and robustness compared to using a single model, achieving a score of 65.1 on Alpaca Eval 2.0, which is higher than GPT-4o. If people are curious, I'll write a more in-depth, technical description. https://lnkd.in/ezP_2ngJ

#Day172 of #500DaysGenerativeAI Read about GPT-4o and implementing RAG on the new model.

To call GPT 4o and inject context, to fine-tune 3.5, or to wait until 4o is available as a fine-tuning model?

Replacement for GPT-3.5-Turbo is here! It's both cheap and performs similarly to Llama 3. Switching over for classification tasks.

This live demo of GPT-4o is incredible: The text-to-speech is miles ahead of ElevenLabs.io https://lnkd.in/gXH8u475

Here’s a quick look at the journey of GPT models, from GPT-1 to GPT-4. Swipe through to see each model’s unique capabilities and how they’ve evolved over time!

After what has felt like a pause in the breathless pace of LLMs, stand by for the next explosion. The demonstrated power of GPT-4o is completely insane. Instant video understanding, seamless conversations, and real time translation. All standard with the free version. Now if only they could make the assistant less obsequious... (Also if you are 'bringing AI to the world, why do you always start with the MAC desktop version?!) https://lnkd.in/ebNEYxVc

LLM Visualization This is actually pretty amazing! It helps to visualize the core components of LLMs like nano-gpt and GPT-3. https://bbycroft.net/llm