📢 Excited to release Gorilla🦍 Gorilla picks from 1000s of APIs to complete user tasks, surpassing even GPT-4! LLMs need to interact with the world through APIs, and Gorilla teaches LLMs APIs. Presenting Gorilla-Spotlight demo🤩 Webpage: gorilla.cs.berkeley.edu

Enhanced Prompting for Text Generation - a prompting framework that encourages the model to detect errors before the generation - effective on various text generation tasks, like summarization, translation, dialogue, and more - applied to 20 datasets using GPT3.5 and GPT-4 and

Статья про GPT-4 с нашими комментариями, спрашивали про размер их модели и в целом влиянии на рынок skillbox.ru/media/code/zak…

Introducing FastChat's OpenAI-compatible API server! Many applications have been built on OpenAI APIs, but now you can effortlessly port them to use open-source alternatives without modifying the code. Blog (with LangChain + Vicuna as an example): lmsys.org/blog/2023-06-0…

Hey everyone, please help we with my request or introduce me to someone who might be interested 🚀 intch.org/t/Vix/ I am using @‌be_intch to connect with the right people to solve my business tasks. Feel free to join my network!

🔥🔥🔥 Introducing the newest WizardLM-70B V1.0 model ! 1. WizardLM-70B V1.0 achieves a substantial and comprehensive improvement on coding, mathematical reasoning and open-domain conversation capacities. 2. This model is license friendly, and follows the same license with Meta

A Survey on LLM-based Autonomous Agents Nice paper presenting a comprehensive survey of LLM-based autonomous agents. It delivers a systematic review of the space and a summary of various applications of LLM-based AI agents in domains like social science and engineering.

LLMs as Optimizers This is a really neat idea. This new paper from Google DeepMind proposes an approach where the optimization problem is described in natural language. An LLM is then instructed to iteratively generate new solutions based on the defined problem and previously

What is multi agent #FractalGPT in under 7 min! Leveraging Ai Agents, Question-Answering agent with NO #hallucinations youtube.com/watch?v=6li4Fg…

Zephyr: Direct Distillation of LM Alignment paper page: huggingface.co/papers/2310.16… We aim to produce a smaller language model that is aligned to user intent. Previous research has shown that applying distilled supervised fine-tuning (dSFT) on larger models significantly improves

Announcing new FractalGPT RAG approach: Общаемся с базой знаний: как мы улучшили точность генеративных ответов LLM с помощью собственного RAG habr.com/ru/articles/77…

🚀Мы запустили агента FractalGPT QA в закрытую бету! FractalGPT QA агент - это вопросно-ответная система, в которую вы можете загружать свои документы или базу знаний компании, задавать по ним вопросы и получать точные и полные ответы. Оставить заявку: forms.yandex.ru/u/63f35bcbeb61…

skillbox.ru/media/code/vy-… Большой обзор про генеративный ИИ - в том числе тренды, опен-сорсные модели, мультимодальные, автономных агентов, прогнозы на 2024й год (и я в том числе от FractalGPT дал там наше мнение)

We have completed our research and released FractalMath - multi-agent system capable to solve arithmetic reasoning tasks with 100% accuracy, reliable, outperforming ChatGPT vc.ru/ml/970699-frac…

LLMs for Mathematical Reasoning Introduces an overview of research developments in LLMs for mathematical reasoning. Discusses advancements, capabilities, limitations, and applications to inspire ongoing research on LLMs for Mathematics.

FractalMath multi-agent Ai system has been accepted on OpenTalks international conference, as first with SOTA accuracy 100% on MultiArith dataset, outperforming ChatGPT with ANY prompting technique (arithmetic reasoning). opentalks.ai/en/timetable20…

We at FractalGPT is moving toward AGI developing multi-agent framework and SDK for autonomous agents. More: fractaltech.ru

Introducing FractalGPT multi-agent Ai technology capable to build cooperative, decentralised Ai landscape. Check video with step by step historical Ai agents review, our agentic RAG system, metrics of efficiency youtube.com/watch?v=Zz59xt…

Вышел наш доклад в сообществе YNDX Family! За 30 минут очень подробно рассказали о возможностях FractalMath - ИИ агентов в решении задачек, которые даже ChatGPT тянет нестабильно. Материал подготовлен на основе онлайн-встречи YNDX Family youtube.com/watch?v=TXyoTn…

🔥 Firefunction-v2, new open-weights function-calling model🔥 I'm super excited to announce Firefunction-v2, our latest open-weights! - Competitive with GPT-4o at function-calling - 1/10 of GPT-4o cost and 2x the speed - Retains both conversation and function-calling