Ollama has released version v0.32.4-rc0, introducing support for the Laguna MLX model. This update includes compatibility with Laguna XS 2, XS 2.1, and S 2.1 variants, implementing a unified quantization policy across dense and MoE layers. The release focuses on optimizing performance by selectively quantizing projections, handling mixed-precision data types, and improving the forward pass with fused operations and cached prefill chunks, all while keeping Laguna weights resident on Metal for efficient GPU access. AI
IMPACT Enhances Ollama's capability to run diverse local models, potentially improving accessibility for users with specific hardware configurations.
RANK_REASON This is a release of a software tool (Ollama) adding support for a specific model variant, not a core AI model release from a frontier lab.
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
