Ahmed Adel’s Post

Chains vs group-chat has been a big differentiator between Flowise and AutoGen and other frameworks. Building logic to progress through a chain and repeat steps if needed adds complexity that is easily overcome by agents collaborating in a group-chat. The downside is that you shift this control from traditional programmatic steps to relying on prompt engineering to make group decisions. For some workflows, this is fine, but others require a more rigid progression of steps (e.g. CI/CD pipelines). A hybrid approach is the best of both worlds.

Last week, with the announcements of GPT-4o and Google I/O, huge bets are on multi-modality agents. Today, we are excited to introduce Multi Agent Flow, powered by LangChain's LangGraph 🕸 Multi agent consists of a team of agents that collaborate together to complete a task delegated by a supervisor. Result is significantly better for long-running task. Here's why: ⚒ Dedicated prompt and tools for each agent 🔄 Reflective loop for auto-correction 🌐 Separate LLMs for different agent Multi Agent Flow supports: - Function Calling LLMs (Claude, Mistral, Gemini, OpenAI) - Multi Modality (image, speech & files coming soon) - API - Prompt input variables Available now in v1.8.0 Repo: https://lnkd.in/dsph3WMU

Efficiency of LLM infrastructures is one of the most important topics for wide scale LLM adoptions and there are many different subtopics such as optimization of inference, optimization of GPU allocations, scaling up architectures etc Here is a very interesting read from Character AI how they optimize their inference infrastructure for their production loads of 20000 qps https://lnkd.in/gzPPzUYZ

Hello, Connections! In today’s rapidly evolving business landscape, the potential applications seem limitless. The rise of large language models has undoubtedly opened new doors, but fine-tuning them comes with its challenges, especially concerning memory limitations. This is where LoRA (Low-Rank Adaptation) steps in, revolutionizing the landscape by addressing these memory issues while preserving model quality. The evolution continues with QLoRA, pushing the boundaries of this technique even further. Take a look at my latest article that delves into this approach, showcasing the fine-tuning of Falcon-7b-Instruct on Google Colab-Free GPU, leveraging LoRA and QLoRA. https://lnkd.in/dgVV5bwU

Watch GPT and Google Gemini go head-to-head in a game of trivia. GPT 5 is reportedly coming this summer, and just last week, Google began making Gemini 1.5 available to all developers. Between those two—plus Claude, Mistral, Llama, Perplexity, and more—it's hard to know which model to use. Sure, you could test them, but human evaluations are expensive. That's why I'm interested in how LLMs themselves can be used for automated evals. In this demo, the questions and assessments are all AI-generated. It's LLMs grading answers given by LLMs to questions written by LLMs. Lots of caveats apply (read the FAQ!), but I had fun building it.

"Through a series of machine learning innovations, we’ve increased 1.5 Pro’s context window capacity far beyond the original 32,000 tokens for Gemini 1.0. We can now run up to 1 million tokens in production. This means 1.5 Pro can process vast amounts of information in one go — including 1 hour of video, 11 hours of audio, codebases with over 30,000 lines of code or over 700,000 words. In our research, we’ve also successfully tested up to 10 million tokens.' https://lnkd.in/ddrFBMnj

This clever approach to efficiently extend context length will help the performance of LLMs for real world applications. I look forward to seeing more exciting advances like this in 2024 to take real-world adoption of open source local LLMs to new heights. #llms #genai

Finally watched this end to end and like everything Andrej Karpathy has put out on YouTube, it's so so good Tokenization explains a large fraction of what makes LLMs do strange things and be bad at stuff we think is "easy". For example, tokenization is the cause of: * why LLMs struggle with basic arithmetic * why they struggle with spelling words or counting letters in words * why you shouldn't end completion requests with a space character I think the future of LLMs and the path towards AGI includes rethinking how we do tokenization — perhaps by figuring out ways to let LLMs and large multimodal models to ingest more of the "raw" experience, closer to the way we experience the world. https://lnkd.in/dRCa7w7r

Best Resources to Learn & Understand Evaluating LLMs via #TowardsAI → https://bit.ly/3WAPbs2

Best Resources to Learn & Understand Evaluating LLMs via #TowardsAI → https://bit.ly/3WAPbs2