Aya 23 is a New State-of-the-Art, Multilingual, Open Weights Model from Cohere
Aya 23 is a research open weights release of an 8 and 35 billion parameter with highly advanced instruction fine-tuned model, covering 23 languages.
Open Source
A collection of 68 posts
Red Hat and IBM Want to Change Open Source AI Development
Open Source AI is Broken. While open source in software fosters collaborative improvement, continuous refinement, and collective ownership, this isn't effectively happening with AI.
Snowflake Launches Arctic: An Open-Source Enterprise-Grade LLM
Snowflake has fully embraced open-source principles with Arctic, releasing it under an Apache 2.0 license and making the model weights and code also available.
Linux Foundation Launches Open Platform for Enterprise AI
This initiative aims to create "hardened," scalable AI systems that leverage open source innovations across a wide spectrum.
AI21 Labs Unveils Jamba: The First Production-Grade Mamba-Based AI Model
Jamba is a groundbreaking SSM-Transformer model that offers the best of both worlds, addressing the drawbacks of traditional Transformer architectures while maintaining their powerful capabilities.
Databricks Open Sources DBRX: A Powerful New Language Model for Enterprises
The weights of the base model (DBRX Base) and the finetuned model (DBRX Instruct) are available on Hugging Face under an open license.
Mistral Announces Le Chat and Mistral Large
Mistral Large achieves strong performance on tests of reasoning, knowledge, coding, math and cross-lingual understanding and is second only to OpenAI's GPT-4
Google Unveils Gemma, A New Family of State-of-the-Art Open Models
Gemma achieved state-of-the-art results across several key benchmarks, outscoring popular open models like Llama-2 and Mistral.
Meta is Teaching AI to Understand and Model the Real World by Watching Videos
Yann LeCun says this approach could lead to more flexible reasoning, planning, and general intelligence.
Cohere For AI Opensources Aya, a 101-Language AI Model
Aya is an open science project that aims to build a multilingual language model via instruction tuning that harnesses the collective wisdom and contributions of people from all over the world.
AI2 Unveils OLMo, the "First Truly Open" Large Language Model
AI2 is not only providing the model weights, but also the training code, actual training data, and an array of associated toolkits, including evaluation tools.
How To Get Started With CodeLlama-70B
CodeLlama-70B-Instruct achieves 67.8 on HumanEval, making it one of the highest performing open models available today.
Stability AI Unveils 1.6B Parameter Stable LM 2 Model
Benchmark results show Stable LM 2 1.6B achieving state-of-the-art results for a model under 2 billion parameters.
OpenVoice is an Open-Source, Instant Voice Cloning AI
The technology stands out for its ability to instantly clone voices with remarkable precision and control.
Mistral Unveils Mixtral 8x7B: A Leading Open SMoE Model
Leveraging the mixture of experts design allows scaling model capacity and performance substantially while controlling memory and compute costs, since any single token does not end up using all weights.
Qiskit, IBM's open-source toolkit for Quantum Computing
The fusion of Watsonx's AI expertise and IBM's quantum computing prowess creates a robust platform for innovation, pushing the boundaries of what's achievable in quantum computing.
Chinese Startup DeepSeek Unveils Impressive New Open Source AI Models
DeepSeek LLM's 67B Base version has demonstrated superior performance compared to the Llama2 70B Base, particularly in areas like reasoning, coding, math, and Chinese comprehension.
OpenAI Delays Launch of GPT Store to Early 2024
Nonetheless, the company says the GPT Store remains firmly on its roadmap. It also hinted at impending upgrades for ChatGPT itself in the near future.
Alibaba Unveils Three New Open-Source AI Models to Advance Research
Qwen-72B was trained on an unprecedented 3 trillion tokens of multi-lingual datasets encompassing text, code, and mathematical content in both English and Chinese.
Google Open-Sources Project Guideline to Enable Visually Impaired Navigation
The company is taking a major step towards improving mobility for the visually impaired by open-sourcing Project Guideline, its computer vision navigation system.
Microsoft Releases New Orca 2 Small Language Model with Enhanced Reasoning Capabilities
Comprehensive benchmarks show Orca 2 significantly outperforming other models of equivalent size on metrics related to language understanding, common sense reasoning, multi-step math problems, reading comprehension, summarization, and more.
Data Provenance Explorer Brings Transparency to AI Data Licensing & Attribution
The interactive platform is designed to offer developers, scholars, and journalists an in-depth look into the composition, lineage, and licensing intricacies of popular AI datasets.
Google Vertex AI Now Supports Mistral AI
Mistral-7B, is now natively available in the Vertex AI Notebooks. This integration provides seamless access to test, fine-tune, and deploy Mistral-7B on Google's managed AI service.
M42 Health Unveils Med42 - A Milestone for Open Access in AI for Healthcare
The model achieved up to 72% accuracy in a zero-shot evaluation on the United States Medical Licensing Examination (USMLE), and outperforms OpenAI's ChatGPT 3.5.