Phi-4 is a 14B parameter, state-of-the-art open model built upon a blend of synthetic datasets, data from filtered public domain websites, and acquired academic books and Q&A datasets. The model was extensively refined and aligned through a combination of supervised fine-tuning and direct preference optimization, ensuring accurate instruction adherence and strong safety protocols. Context length: 16k … Read more
IBM’s updated Granite 3.1 models are here! 4 sizes using two different architectures. Mixture of expert models are designed for low latency usage. Dense models are designed for tool-based use cases. IBM is also releasing their embedding models today! English only: Multilingual:
New state of the art 70B model. Llama 3.3 70B offers similar performance compared to Llama 3.1 405B model. Meta Llama 3.3: A State-of-the-Art Multilingual AI Model Meta introduces its latest breakthrough, the Llama 3.3, a cutting-edge 70B multilingual large language model (LLM) that rivals the performance of the Llama 3.1 405B model. Key Features … Read more
Snowflake launched a new state-of-the-art embedding model! Arctic Embed 2 Run: ollama pull snowflake-arctic-embed2. Snowflake is excited to announce the release of Arctic Embed 2.0, the next iteration of our frontier embedding models, which now empower multilingual search. While our previous releases have been well received by our customers, partners and the open source community, … Read more
Qwen QwQ is an experimental research model focused on advancing AI reasoning capabilities. QwQ-32B-Preview is an experimental model from the Qwen Team, designed to explore advanced AI reasoning capabilities. While it shows potential, it is still in the research phase and comes with several notable limitations. Key Challenges QwQ demonstrates remarkable performance across these benchmarks: … Read more
The latest update to the Ollama Python library introduces several powerful features: Getting Started Begin by installing or updating the Ollama Python library: bashCopy codepip install -U ollama Passing Python Functions as Tools 1. Define a Python Function Create a standard Python function, including type annotations for parameters and return values. Adding a Google-style docstring … Read more
The latest series of Code-Specific Qwen models, with significant improvements in code generation, code reasoning, and code fixing. Qwen 2.5 Coder series of models are now updated in 6 sizes: 0.5B, 1.5B, 3B, 7B, 14B and 32B. There are significant improvements in code generation, code reasoning and code fixing. The 32B model has competitive performance with OpenAI’s GPT-4o. Qwen2.5 Coder … Read more
The latest October 2024 update to the AI Toolkit in Visual Studio Code is a game-changer for developers, researchers, and AI enthusiasts. Packed with groundbreaking features, this release revolutionizes how users interact with generative AI models, offering unparalleled flexibility, performance, and accessibility. Among the most exciting updates: support for Ollama models. Unlocking Multi-Model Power in VS … Read more
This experimental pre-release brings initial support for Llama 3.2 Vision. After downloading the Ollama 0.4 pre-release, you can launch it with: Before the full release of 0.4, we aim to further enhance inference quality and performance and test Ollama’s new Go model runtime. Testing on different platforms and model architectures is welcomed and greatly appreciated. … Read more
SmolLM2 is a family of compact language models available in three size: 135M, 360M, and 1.7B parameters. SmolLM2 is a series of compact language models offered in three sizes: 135M, 360M, and 1.7B parameters. These models are designed to handle a broad spectrum of tasks while remaining lightweight enough to operate directly on-device. Benchmark results … Read more
