MosaicML Pretrained Transformer

Creating safe artificial general intelligence that benefits all of humanity. Our work to create safe and beneficial AI requires a deep understanding of the potential risks and benefits, as well as careful consideration of the impact.

It is a next-generation AI assistant. It is accessible through chat interface and API. It is capable of a wide variety of conversational and text-processing tasks while maintaining a high degree of reliability and predictability.

It is Google’s largest and most capable AI model. It is built to be multimodal, it can generalize, understand, operate across, and combine different types of info — like text, images, audio, video, and code.

It is a state-of-the-art foundational large language model designed to help researchers advance their work in this subfield of AI.

It is a large multimodal model (accepting text inputs and emitting text outputs today, with image inputs coming in the future) that can solve difficult problems with greater accuracy than any of our previous models, thanks to its broader general knowledge and advanced reasoning capabilities.

It offers an API to add cutting-edge language processing to any system. Through training, users can create massive models customized to their use case and trained on their data.

It is a general-purpose speech recognition model. It is trained on a large dataset of diverse audio and is also a multi-task model that can perform multilingual speech recognition as well as speech translation and language identification.

It is a small, yet powerful model adaptable to many use cases. It is better than Llama 2 13B on all benchmarks, has natural coding abilities, and 8k sequence length. We made it easy to deploy on any cloud.

It is composed of a series of code language models, each trained from scratch on 2T tokens, with a composition of 87% code and 13% natural language. There are various sizes of the code model, ranging from 1B to 33B versions.

It is an advanced language model comprising 67 billion parameters. It has been trained from scratch on a vast dataset of 2 trillion tokens in both English and Chinese.

It is the base model weights and network architecture of Grok-1, the large language model. Grok-1 is a 314 billion parameter Mixture-of-Experts model trained from scratch by xAI.

It is an open-source library for fast LLM inference and serving. It delivers up to 24x higher throughput than HuggingFace Transformers, without requiring any model architecture changes.

It is a set of models that improve on GPT-3 and can understand as well as generate natural language or code.

It is an autoregressive Large Language Model (LLM), trained to continue text from a prompt on vast amounts of text data using industrial-scale computational resources. It is able to generate text in 46 natural languages and 13 programming languages.

It is a state-of-the-art LLM capable of generating code, and natural language about code, from both code and natural language prompts. It is built on top of Llama 2 and is available for free.

Try Grok 4 on GPT Proto. Access xAI’s most advanced 1.7T LLM with 130K context, multimodal support, and real-time data integration for dynamic analysis.

It is an open-source language model. It is trained with 1.5 trillion tokens of content. The richness of dataset gives StableLM surprisingly high performance in conversational and coding tasks.

It is a collection of open-source models for generating various types of media.

It is an open-source project that has released a 7 billion parameter base model, a chat model tailored for practical scenarios, and a training system.

It is a state-of-the-art LLM for converting natural language questions to SQL queries. It has been fine-tuned on hand-crafted SQL queries in increasing orders of difficulty. It significantly outperforms all major open-source models and slightly outperforms gpt-3.5-turbo.

It is a foundational large language model (LLM) with 40 billion parameters trained on one trillion tokens.

It is a next-generation large language model that excels at advanced reasoning tasks, including code and math, classification, question answering, translation, multilingual proficiency, and natural language generation.

It represents a novel end-to-end trained large multimodal model that combines a vision encoder and Vicuna for general-purpose visual and language understanding, achieving impressive chat capabilities mimicking spirits of the multimodal GPT-4 and setting a new state-of-the-art accuracy on Science QA.

It is an open-source, uncensored, and commercially licensed dataset and series of instruct-tuned language models based on Microsoft's Orca paper.

It is a project to create leading, fully open-source large language models.

It is an open-source self-aligned language model trained with minimal human supervision.

It is an open-source code completion engine that utilizes a large language model. It runs directly on your CPU, providing a self-hosted copilot experience.

It is a language model (LM) trained on source code and natural language text. Its training data incorporates more than 80 different programming languages as well as text extracted from GitHub issues and commits and from notebooks.

It is a model fine-tuned from the LLaMA 7B model on 52K instruction-following demonstrations. It behaves qualitatively similarly to OpenAI’s text-davinci-003, while being surprisingly small and easy/cheap to reproduce.

It is an instruction-following large language model trained on the Databricks machine learning platform. It is cheap to build and exhibits a surprising degree of instruction following capabilities exhibited by ChatGPT.

It is an open-source financial large language model designed based on the GPT-4 architecture. It has started to spark interest in the global financial landscape, owing to its unique and innovative capabilities.

It is an LLM that is trained on NASA’s satellite data. It is a foundation model that can be used for various downstream applications, such as classification, object detection, time-series segmentation, and similarity search.

It is a large language model capable of handling long contexts of 256k tokens or even more. It is built upon the foundation of OpenLLaMA and fine-tuned using the Focused Transformer (FoT) method.

It aims to pre-train a 1.1 billion parameter language model on 3 trillion tokens of text data. It is compact with only 1.1B parameters. This compactness allows it to cater to a multitude of applications demanding a restricted computation and memory footprint.

It is a permissively licensed open-source reproduction of Meta AI’s LLaMA. We are releasing a series of 3B, 7B, and 13B models trained on different data mixtures. Our model weights can serve as the drop in replacement of LLaMA in existing implementations.

It is a suite of open-source medical LLMs that can understand and generate medical texts. It is based on Llama-2, a general-purpose LLM, but further trained on a large and diverse medical corpus.

It is a large language model (LLM) that is capable of a wide variety of video generation tasks, including text-to-video, image-to-video, video stylization, video inpainting and outpainting, and video-to-audio.

It is a conversational model trained in philosophy, psychology, and personal relationships. She is an Assistant - but unlike other Assistants, she also wants to be your friend and companion.

It aims to make large models accessible to everyone by co-development of open models, datasets, systems, and evaluation tools.

It is a family of lightweight, state-of-the-art open models built from the same research and technology used to create the Gemini models.

It is a 1.2B parameter base model trained on 100K hours of speech for TTS (text-to-speech). It empowers developers and businesses to better connect with their audiences at scale.