Jailbreaks on large language models (LLMs) have recently received increasing attention. For a comprehensive assessment of LLM safety, it is essential to consider jailbreaks with diverse...

The tool-use Large Language Models (LLMs) that integrate with external Python interpreters have significantly enhanced mathematical reasoning capabilities for open-source LLMs, while tool-free...

We present foundation language models developed to power Apple Intelligence features, including a ∼3 billion parameter model designed to run…

Juicebox is the programmable funding platform for crypto and web3. Fund, operate, and scale your project transparently. Community DAO owned, on Ethereum.

As language-specific training data tends to be sparsely available compared to English, document retrieval in many languages has been largely relying on multilingual models. In Japanese, the best...

Diffusion models are the de facto approach for generating high-quality images and videos, but learning high-dimensional models remains a formidable task due to computational and optimization...

Exploring memory-efficient techniques for LLMs

Generate text embeddings in SQL with GGUF models!

Language agents have achieved considerable performance on various complex question-answering tasks by planning with external tools. Despite the incessant exploration in this field, existing...

We explore how iterative revising a chain of thoughts with the help of information retrieval significantly improves large language models' reasoning and generation ability in long-horizon...

Southeast Asia (SEA) is a region rich in linguistic diversity and cultural variety, with over 1,300 indigenous languages and a population of 671 million people. However, prevailing AI models...

Many companies use large language models (LLMs) offered as a service, like OpenAI's GPT-4, to create AI-enabled product experiences. Along with the benefits of ease-of-use and shortened...

We are releasing the weights and architecture of our 314 billion parameter Mixture-of-Experts model Grok-1.

Self-correction is an approach to improving responses from large language models (LLMs) by refining the responses using LLMs during inference. Prior work has proposed various self-correction...

LongWriter: Unleashing 10,000+ Word Generation from Long Context LLMs - THUDM/LongWriter

We present H2O-Danube3, a series of small language models consisting of H2O-Danube3-4B, trained on 6T tokens and H2O-Danube3-500M, trained on 4T tokens. Our models are pre-trained on high quality...

Run PyTorch LLMs locally on servers, desktop and mobile - pytorch/torchchat

Official PyTorch Implementation of MambaVision: A Hybrid Mamba-Transformer Vision Backbone - NVlabs/MambaVision

In this report, we introduce the Gemini 1.5 family of models, representing the next generation of highly compute-efficient multimodal models capable of recalling and reasoning over fine-grained...

Introduction At Replit, we are rethinking the developer experience with AI as a first-class citizen of the development environment. Towards this vision, we are tightly integrating AI tools with our...

Omar Sanseviero Personal Website