@meta
  v: 1
  route: /runtime/ollama
  generated: 2026-06-10T09:08:08.689Z

@intent
  purpose:    Install, configure, and serve LLMs with Ollama.
  audience:   self-hoster, ai-engineer, mac-user, devops-engineer
  capability: install, serve_local_llms, compare_runtimes, open_external_docs

@state
  slug: ollama
  name: Ollama
  family: cross-platform
  type: cli
  license: MIT
  primary_platform: macOS · Linux · Windows · Docker
  platforms[4]: macOS, Linux, Windows, Docker
  model_formats[2]: GGUF, safetensors
  api_compatibility[2]: REST :11434, OpenAI-compatible adapters
  install_command: curl -fsSL https://ollama.com/install.sh | sh
  install_secondary: irm https://ollama.com/install.ps1 | iex
  homepage_url: https://ollama.com
  github_url: https://github.com/ollama/ollama
  docs_url: https://github.com/ollama/ollama/blob/main/docs/README.md
  feature_count: 4
  feature_labels[4]: Model registry, REST + SDKs, Integrations, GGUF + safetensors
  best_for[4]: First-time local LLM users, Quick model swapping via `ollama run <name>`, "Cross-platform setups (same workflow on Mac, Linux, Windows)", "Integrations: Claude Code, OpenClaw, Codex, Copilot CLI all read Ollama on :11434"
  caveats[3]: Single-tier in-memory KV cache; mid-session context shifts cause recomputation, No paged attention; throughput trails vLLM under heavy concurrency, GUI is minimal; pair with Open WebUI or one of the community clients for chat

@actions
  - id: open_homepage
    method: GET
    href: https://ollama.com
  - id: open_github
    method: GET
    href: https://github.com/ollama/ollama
  - id: open_docs
    method: GET
    href: https://github.com/ollama/ollama/blob/main/docs/README.md
  - id: view_index
    method: GET
    href: /runtime/
  - id: view_calculator
    method: GET
    href: /#calculator

@context
  > Single-binary CLI with a built-in model registry. Curl-install, run a model in one command, hit it via REST on port 11434. Backed by llama.cpp under the hood; supports GGUF and safetensors. The default starting point for almost everyone running local LLMs.

@nav
  self:      /runtime/ollama
  parents:   [/, /runtime/]
  peers:     [/runtime/lm-studio, /runtime/vllm, /runtime/mlx, /runtime/omlx]
  drilldown: /#calculator