InfoQ Homepage Large language models Content on InfoQ
-
DeepSeek Open-Sources DeepSeek-V3, a 671B Parameter Mixture of Experts LLM
DeepSeek open-sourced DeepSeek-V3, a Mixture-of-Experts (MoE) LLM containing 671B parameters. It was pre-trained on 14.8T tokens using 2.788M GPU hours and outperforms other open-source models on a range of LLM benchmarks, including MMLU, MMLU-Pro, and GPQA.
-
Google Releases Experimental AI Reasoning Model
Google has introduced Gemini 2.0 Flash Thinking Experimental, an AI reasoning model available in its AI Studio platform.
-
Google Vertex AI Provides RAG Engine for Large Language Model Grounding
Vertex AI RAG Engine is a managed orchestration service aimed to make it easier to connect large language models (LLMs) to external data sources to be more up-to-date, generate more relevant responses, and hallucinate less.
-
Android Studio Ladybug Update Adds Gemini Support, New Debugging Features, and More
In its recent update to Android Studio Ladybug (2024.2.2), Google has added new Gemini Code Transforms to modify, refactor, or create code, debugging and testing tools, and developer experience improvements. Additionally, the IDE adopts the latest IntelliJ 2024.2 platform release.
-
Major LLMs Have the Capability to Pursue Hidden Goals, Researchers Find
Researchers at AI safety firm Apollo Research found that AI agents may covertly pursue misaligned goals and hide their true objectives. Known as in-context scheming, this behavior does not seem to be accidental as LLMs explicitly reason about deceptive strategies and consider them a viable strategy.
-
HuatuoGPT-o1: Advancing Complex Medical Reasoning with AI
Researchers from The Chinese University of Hong Kong, Shenzhen, and the Shenzhen Research Institute of Big Data have introduced HuatuoGPT-o1, a medical large language model (LLM) designed to improve reasoning in complex healthcare scenarios.
-
Google Releases PaliGemma 2 Vision-Language Model Family
Google DeepMind released PaliGemma 2, a family of vision-language models (VLM). PaliGemma 2 is available in three different sizes and three input image resolutions and achieves state-of-the-art performance on several vision-language benchmarks.
-
Nvidia Announces Arm-Powered Project Digits, Its First Personal AI Computer
Capable of running 200B-parameter models, Nvidia Project Digits packs the new Nvidia GB10 Grace Blackwell chip to allow developers to fine-tune and run AI models on their local machines. Starting at $3,000, Project Digits targets AI researchers, data scientists, and students to allow them to create their models using a desktop system and then deploy them on cloud or data center infrastructure.
-
Nvidia Nemotron Models Aim to Accelerate AI Agent Development
Nvidia has launched Llama Nemotron large language models (LLMs) and Cosmos Nemotron vision language models (VLMs) with a special emphasis on workflows powered by AI agents such as customer support, fraud detection, product supply chain optimization, and more. Models in the Nemotron family come in Nano, Super, and Ultra sizes to better fit the requirements of diverse systems.
-
Meta Open-Sources Byte Latent Transformer LLM with Improved Scalability
Meta open-sourced Byte Latent Transformer (BLT), a LLM architecture that uses a learned dynamic scheme for processing patches of bytes instead of a tokenizer. This allows BLT models to match the performance of Llama 3 models but with 50% fewer inference FLOPS.
-
Hugging Face Smolagents is a Simple Library to Build LLM-Powered Agents
Smolagents is a library created at Hugging Face to build agents based on large language models (LLMs). Hugging Faces says its new library aims to be simple and LLM-agnostic. It supports secure "agents that write their actions in code" and is integrated with Hugging Face Hub.
-
LLaMA-Mesh: NVIDIA’s Breakthrough in Unifying 3D Mesh Generation and Language Models
NVIDIA researchers have introduced LLaMA-Mesh, a groundbreaking approach that extends large language models (LLMs) to generate and interpret 3D mesh data in a unified, text-based framework. LLaMA-Mesh tokenizes 3D meshes as plain text, enabling the seamless integration of spatial and textual information.
-
DeepThought-8B Leverages LLaMA-3.1 8B to Create a Compact Reasoning Model
DeepThought-8B is a small "reasoning" model built on LLaMA-3.1 8B that can carry through decision-making processes step by step, similarly to how OpenAI o1 does but in a much smaller package.
-
Qwen Team Unveils QwQ-32B-Preview: Advancing AI Reasoning and Analytics
Qwen Team introduced QwQ-32B-Preview, an experimental research model designed to improve AI reasoning and analytical capabilities. Featuring a 32,768-token context and cutting-edge transformer architecture, it excels in math, programming, and scientific benchmarks like GPQA and MATH-500. Available on Hugging Face, it invites researchers to explore its features and contribute to its development.
-
InstaDeep Open-Sources Genomics AI Model Nucleotide Transformers
Researchers from InstaDeep and NVIDIA have open-sourced Nucleotide Transformers (NT), a set of foundation models for genomics data. The largest NT model has 2.5 billion parameters and was trained on genetic sequence data from 850 species. It outperforms other state-of-the-art genomics foundation models on several genomics benchmarks.