News Hub

Meta released Muse Glimmer, a 30B multimodal model under Apache 2.0 license, now available on Ollama for local coding agents.

Ollama raised 88 million dollars and serves 8.9 million developers of open models.

Gemma 4 runs up to 90% faster on Apple Silicon in Ollama 0.31 using multi-token prediction with MLX.

Ollama's MLX engine updated for Apple Silicon delivers faster responses, higher quality outputs, and reduced memory usage.

Ollama 0.30 improves performance and GGUF model compatibility via llama.cpp, expanding hardware support.

NVIDIA Nemotron 3 Ultra is designed for high-throughput reasoning and long-running agent workflows.

OpenJarvis v1.0 is an open-source personal AI agent framework running on local hardware.

Ollama now supports MLX on Apple Silicon for faster inference in preview.

Ollama enables OpenClaw setup in under two minutes using a single command.

Ollama added subagents and web search capabilities to Claude Code.

OpenClaw is a personal AI assistant connecting messaging apps to local AI coding agents.

Ollama launch command sets up coding tools like Claude Code with local or cloud models without configuration.

Ollama added experimental local image generation support on macOS.

Ollama now supports Anthropic Messages API, enabling Claude Code compatibility with open models.

Ollama integrated with OpenAI Codex CLI to run open-weight code models.

Ollama partners with OpenAI and ROOST to release gpt-oss-safeguard models for safety classification under Apache 2.0.

MiniMax M2 model available on Ollama for coding and agentic workflows.

Ollama tests performance on NVIDIA DGX Spark hardware.

Ollama added GLM-4.6 and Qwen3-coder models with improved tool calling capabilities.

Ollama now supports Alibaba's Qwen3-VL multimodal model.