Subscribe

How to install ollama (x86_64)

Get up and running with large language models locally

Install

sudo yum -y install https://extras.getpagespeed.com/release-latest.rpm
sudo yum -y install ollama

Description

Ollama allows you to run large language models locally. Get up and running with Llama 3, Mistral, Gemma, and other open source models.

RPMs

Danila Vershinin (2026-03-19) - Added checks for npm and git installation before OpenClaw setup. · - Improved local performance of Claude Code by preventing cache issues. · - Fixed support for launching OpenClaw models correctly. · - Registered Ollama's websearch package for OpenClaw.
Danila Vershinin (2026-03-18) - Added web search and fetch plugin for OpenClaw. · - Local models can now search the web for content. · - Non-interactive mode for `ollama launch` introduced. · - Improved benchmarking tool in `./cmd/bench`. · - `ollama launch openclaw` now uses official auth provider. · - Skip `--install-daemon` if systemd is unavailable.
Danila Vershinin (2026-03-15) - Improved model ordering for better performance. · - Cloud models now connect automatically without downloading. · - Support for setting compaction window in Claude Code.
Danila Vershinin (2026-03-07) - Improved API interpretation for thinking levels like 'medium'. · - Added context length support for compaction in 'ollama launch'.
Danila Vershinin (2026-03-04) - Fixed glm-ocr image tags in renderer prompts. · - Improved stability of model rendering functionality. · - Enhanced compatibility with recent dependencies. · - Addressed minor bugs affecting user experience. · - Updated documentation for clarity and accuracy.
Danila Vershinin (2026-03-02) - Added compatibility for imported GGUF models. · - Improved model handling and performance. · - Fixed various bugs affecting model imports. · - Enhanced error messages for better user feedback. · - Updated documentation for new features.
Danila Vershinin (2026-02-27) - Added stable tool call indexing for glm47 and qwen3 parsers. · - Improved model parsing performance and reliability. · - Enhanced compatibility with existing model formats. · - Fixed minor bugs in tool call handling. · - Updated documentation for new features and usage.
Danila Vershinin (2026-02-26) - Initial package (CPU-only)