The latest Gemini-Exp-1114 went straight to the top of the Imarena Chatbot Arena. It excels across technical and creative domains: - Overall #3 -> #1 - Math: #3 -> #1 - Hard Prompts: #4 -> #1 - Creative Writing #2 -> #1 - Vision: #2 -> #1 - Coding: #5 -> #3 - Overall (StyleCtrl): #4 -> #4 #Gemini Google https://lnkd.in/g8KJDQ2B
Bikash Koley’s Post
More Relevant Posts
-
Product and UX Design at Google • Generative AI • Conversational AI • AI/ML products • Entrepreneur • Advisor • xApple, xAutodesk, Agencies — Opinions & Comments are my own.
New Gemini model available, Gemini 1114 in AI Studio. I recently used it to analyze unstructured post-it notes from a workshop. With that Gemini could provide themes and sentimate. It returned transcribed quotes of what it saw written in the response, which helped with validation. From there I was able to determine the number of instances something was written down and generate a dataset that could be turned into a chart. Great vision capability… Data prep to report, 2 hours. #gemini #ai
The latest Gemini-Exp-1114 went straight to the top of the Imarena Chatbot Arena. It excels across technical and creative domains: - Overall #3 -> #1 - Math: #3 -> #1 - Hard Prompts: #4 -> #1 - Creative Writing #2 -> #1 - Vision: #2 -> #1 - Coding: #5 -> #3 - Overall (StyleCtrl): #4 -> #4 #Gemini Google https://lnkd.in/g8KJDQ2B
Google drops new Gemini model and it goes straight to the top of the LLM leaderboard
tomsguide.com
-
Hi #everyone 🖐 Here is another interesting Large multimodal model, PaliGemma 💥 PaliGemma is an open-source Large Multimodal Model (LMM) released by Google last week. You can use it for Visual Question Answering (VQA), object detection, or image segmentation. PaliGemma 3B is available in three different versions, differing in input image resolution (224, 448, and 896) and input text sequence length (128, 512, and 512 tokens, respectively). Website: https://lnkd.in/gBkvU_eR Git Repo: https://lnkd.in/gbNvugXs #opensource #llm #objectdetection #generativeai
Gemma - a family of lightweight, state-of-the art open models from Google | Google for Developers
ai.google.dev
-
I'm elated to share with you all a project that's been a labor of love and intellect: aiStoryGenerator. This project represents a leap into the depths of AI, harnessing the power of OpenAI's latest advancements to turn imaginative concepts into captivating narratives. What's new with aiStoryGenerator: 🚀 Leveraged OpenAI's GPT-3.5-turbo for dynamic and imaginative story creation. 🎨 Features a sleek, React-driven interface, offering a seamless bridge between human creativity and AI's potential. 📚 Allows the creation of diverse narratives, from epic adventures to personalized bedtime stories, all tailored to your input. 🔍 Fine-tuned error handling to ensure a seamless narrative flow in user interactions. The aiStoryGenerator not only exemplifies my enthusiasm for AI's transformative impact on our daily lives but also underscores my commitment to pushing the envelope in software solutions. Take a moment to immerse yourself in the potential of AI-driven narratives. Visit the project on my GitHub and let your imagination run wild: https://lnkd.in/ekVnxQcT In an effort to expand the project's horizon, I pose a question to this brilliant community: Should we integrate text-to-speech to give voice to our tales, or should we explore visual storytelling with OpenAI's DALL·E to illustrate our stories? Your thoughts and feedback will help shape the future of narrative AI. Let's create, share, and innovate together! #AI #OpenAI #Storytelling #GPT3 #ReactJS #SoftwareDevelopment #TechInnovation #CreativeCoding #MachineLearning #DALLE #PersonalizedStories #DigitalNarratives #TechCommunity #AIforEveryone
GitHub - dtyutyunik/aiStoryGenerator
github.com
-
Senior Machine Learning Engineer at Jobiak & M.Tech (CSE), IIT Guwahati | Ex GTO R&D Cognizant
Best way to understand the model capabilities on different Benchmark with figures.
Chatbot arena rankings are out - Sonnet didn't quite unseat GPT-4o across most categories, but did score the highest on coding - a really nice achievement considering it isn't even the biggest of the Claude 3 model family. We are expecting Opus 3.5 in a few months' time and it would be very interesting to see how it does. Apart from Sonnet 3.5, it was interesting to see how other models did. It is notable how weaker Google models are at coding, which is a bit of a downside for use cases that might take advantage of the long context window (e.g. putting the whole codebase in). It is also a bit surprising how much lower down Opus is - it is the most expensive model by far and I hope they will pull of a miracle with Opus 3.5 to justify the price. Note that while the difference might look dramatic, this was done for the sake of comparison, the margins in reality are very tight at the top - a difference of a few percentage points at most. Have a look at the full rankings here: https://lnkd.in/eqYG4K43 and try out the chatbot arena - btw it now includes image upload and vision capability is where Sonnet 3.5 exceeds GPT-4o in my limited testing.
-
-
Engineering Leader Building GenAI Solutions | Cloud Architecture & RAG Systems | AWS, Lululemon, Nike
🎉 New release of LLM Bridge I do a bit of nights and weekends work with Rust, leveraging it for GenAI projects. Recently I added tool use support (also known as function calling) to my Rust SDK for interacting with Large Language Model APIs. 📢 New Feature: I've added support for tool use (also known as function calling), allowing you to define custom tools that language models can utilize during conversations. 🔧 Key updates: - Define custom tools with parameters - Seamless integration with both OpenAI and Anthropic APIs - Easy-to-use builder pattern for tool creation 👨💻 As always, LLM Bridge aims to provide a simple yet powerful interface for working with LLMs in Rust. I continually work to improve and expand its capabilities as I utilize it in my own projects. 🙏 Love to hear your thoughts! Any suggestions for future improvements? Your feedback is invaluable as I continue to develop LLM Bridge. Check out the latest release: https://lnkd.in/gJEWnqFs #RustLang #LLM #AI #OpenSource #DeveloperTools
-
-
Beginners Guide to GPT4 API & ChatGPT 3.5 Turbo API Tutorial
Beginners Guide to GPT4 API & ChatGPT 3.5 Turbo API Tutorial - AIPressRoom
https://aipressroom.com
-
🎓 Keen Learner | 🔧 Technology Aficionado | 🤝 Philanthropic Volunteer | ♟️ Strategic Chess Player | 💡 Innovating for Impact
🚀 Exciting News! 🚀 I'm thrilled to share my latest project: Chat BOT using LLMs! 🤖✨ 🔍 Project Overview: I've built an intelligent chatbot leveraging Large Language Models (LLMs) and Google's Gemini Pro generative AI. This innovative project uses Streamlit for an interactive and user-friendly interface. 🔧 Tech Stack: - Streamlit for a seamless and responsive UI - Google Gemini Pro for powerful language generation - Python for smooth integration and functionality 💡 Features: - Real-time Q&A: Ask any question and get instant, insightful responses. - Session History: Easily track and review your past interactions. - Dynamic Interface: A clean, intuitive layout that enhances user experience. 🌟 How It Works: Enter your question in the text input. Hit 'Submit' and watch the magic happen! Check the sidebar for a history of your interactions. 🔗 Stay tuned for more updates and feel free to reach out if you have any questions or feedback! 🌐 GitHub repo - https://lnkd.in/gmB6PFU3 #AI #MachineLearning #ChatBot #GoogleAI #Streamlit #TechInnovation #PythonProjects #LLM
-
I write about the emerging conflict between AI and knowledge work. Follow me to see how it all unfolds!
While some might view this as a “loss”, the sheer difference in model size and inference cost (and deployment flexibility) between Sonnet and the OpenAI flagship is astounding. Sonnet should be everyone’s current default model for scaled use cases that need gpt-4 level capabilities right now.
Chatbot arena rankings are out - Sonnet didn't quite unseat GPT-4o across most categories, but did score the highest on coding - a really nice achievement considering it isn't even the biggest of the Claude 3 model family. We are expecting Opus 3.5 in a few months' time and it would be very interesting to see how it does. Apart from Sonnet 3.5, it was interesting to see how other models did. It is notable how weaker Google models are at coding, which is a bit of a downside for use cases that might take advantage of the long context window (e.g. putting the whole codebase in). It is also a bit surprising how much lower down Opus is - it is the most expensive model by far and I hope they will pull of a miracle with Opus 3.5 to justify the price. Note that while the difference might look dramatic, this was done for the sake of comparison, the margins in reality are very tight at the top - a difference of a few percentage points at most. Have a look at the full rankings here: https://lnkd.in/eqYG4K43 and try out the chatbot arena - btw it now includes image upload and vision capability is where Sonnet 3.5 exceeds GPT-4o in my limited testing.
-
-
Project : RAG Bot using Llamaindex to answer queries from your PDF Documents Project Highlights: 🔍 Llamaindex Integration: Supercharge your document search and retrieval experience by tapping into the dynamic capabilities of Llamaindex. 🌈 Chroma DB: Seamlessly manage and organize PDF documents for optimal accessibility. 🔥 LLAMA CPP Model: Benefit from state-of-the-art open source NLP model for accurate and intelligent responses. The main goal of llama.cpp is to run the LLaMA model using 4-bit integer quantization on a MacBook Pro #rag #github #llamaindex #chroma #llama #gpt #largelanguagemodels #nlp #ai #ml
GitHub - hk3427/RAG-KnowledgeLLM-LlamaIndex-BOT: LLM Chatbot with Retrieval Augmented Generation using Llamaindex. It works both in online and offline mode.
github.com
-
Helpful visualization of LLM comparisons:
Chatbot arena rankings are out - Sonnet didn't quite unseat GPT-4o across most categories, but did score the highest on coding - a really nice achievement considering it isn't even the biggest of the Claude 3 model family. We are expecting Opus 3.5 in a few months' time and it would be very interesting to see how it does. Apart from Sonnet 3.5, it was interesting to see how other models did. It is notable how weaker Google models are at coding, which is a bit of a downside for use cases that might take advantage of the long context window (e.g. putting the whole codebase in). It is also a bit surprising how much lower down Opus is - it is the most expensive model by far and I hope they will pull of a miracle with Opus 3.5 to justify the price. Note that while the difference might look dramatic, this was done for the sake of comparison, the margins in reality are very tight at the top - a difference of a few percentage points at most. Have a look at the full rankings here: https://lnkd.in/eqYG4K43 and try out the chatbot arena - btw it now includes image upload and vision capability is where Sonnet 3.5 exceeds GPT-4o in my limited testing.
-
Corporate training | Consultant |Product Development (Currently Hiring Embedded professionals )
1wInteresting