Ollama
Use Ollama when you want the quickest verified path from benchmark result to a real local run. It is the best default starting point for most people.
What's My AIChoose the right local runtime after you get an American-model result.
Guide
guide
What to use after WhatsMy.AI says your machine is ready for an American local model.
The result page recommends a runtime because setup style matters. Some people want the fastest verified pull, others want a GUI, and some care most about raw local tuning.
LM Studio
Use LM Studio when you want a graphical model browser and a lower-friction way to pull a tracked GGUF build without living in the terminal.
llama.cpp
Use llama.cpp when you care more about low-level tuning, GGUF control, and a flexible local stack than convenience.
runtime smoke
Monthly runtime smoke matrix
Each row installs or updates the tracked runtime, downloads the starter model, and proves one local inference with the pinned prompt bundle.
These rows use hosted CPU runners so stale guidance is visible before the public install copy drifts too far from reality.
Ollama
Runtime guidance currently needs review
Last verified: Jun 1, 2026, 3:32 PM
Tested runtime version: Warning: could not connect to a running Ollama instance
Monthly smoke cadence (31-day review window)
Prompt bundle: 2026.03-reference-lm-prompts-v1
Linux
GitHub-hosted Ubuntu x64 CPU runner
Install recipe: Install the latest standalone Ollama CLI asset for Linux before each run.
Last verified: Jun 1, 2026, 3:32 PM
Tested version: Warning: could not connect to a running Ollama instance
Model pull: Granite 4.0 Micro
Status: verified
macOS
GitHub-hosted macOS CPU runner
Install recipe: Install the latest standalone Ollama CLI asset for macOS before each run.
Last verified: Not yet verified
Tested version: Not yet verified
Model pull: Granite 4.0 Micro
Stale: Latest smoke run failed during local inference.
Windows
GitHub-hosted Windows x64 CPU runner
Install recipe: Install the latest standalone Ollama CLI asset for Windows before each run.
Last verified: Jun 1, 2026, 3:32 PM
Tested version: Warning: could not connect to a running Ollama instance
Model pull: Granite 4.0 Micro
Status: verified
LM Studio
Runtime guidance currently needs review
Last verified: Jun 1, 2026, 3:32 PM
Tested runtime version: [38;5;166m __ __ ___ ______ ___ _______ ____[0m
Monthly smoke cadence (31-day review window)
Prompt bundle: 2026.03-reference-lm-prompts-v1
Linux
GitHub-hosted Ubuntu x64 CPU runner
Install recipe: Install the latest LM Studio headless runtime (`llmster`) with the official Unix installer.
Last verified: Jun 1, 2026, 3:32 PM
Tested version: [38;5;166m __ __ ___ ______ ___ _______ ____[0m
Model pull: Granite 4.0 Micro
Status: verified
macOS
GitHub-hosted macOS CPU runner
Install recipe: Install the latest LM Studio headless runtime (`llmster`) with the official Unix installer.
Last verified: Not yet verified
Tested version: Not yet verified
Model pull: Granite 4.0 Micro
Stale: Latest smoke run failed during local inference.
Windows
GitHub-hosted Windows x64 CPU runner
Install recipe: Install the latest LM Studio headless runtime (`llmster`) with the official PowerShell installer.
Last verified: Not yet verified
Tested version: Not yet verified
Model pull: Granite 4.0 Micro
Stale: Latest smoke run failed during install/update.
llama.cpp
Runtime guidance currently needs review
Last verified: Not yet verified
Tested runtime version: Not yet verified
Monthly smoke cadence (31-day review window)
Prompt bundle: 2026.03-reference-lm-prompts-v1
Linux
GitHub-hosted Ubuntu x64 CPU runner
Install recipe: Install the latest llama.cpp prebuilt CPU release for Ubuntu before each run.
Last verified: Not yet verified
Tested version: Not yet verified
Model pull: Granite 4.0 Micro GGUF
Stale: Latest smoke run failed during artifact collection.
macOS
GitHub-hosted macOS CPU runner
Install recipe: Install the latest llama.cpp prebuilt binary release for macOS before each run.
Last verified: Not yet verified
Tested version: Not yet verified
Model pull: Granite 4.0 Micro GGUF
Stale: Latest smoke run failed during local inference.
Windows
GitHub-hosted Windows x64 CPU runner
Install recipe: Install the latest llama.cpp prebuilt CPU release for Windows before each run.
Last verified: Not yet verified
Tested version: Not yet verified
Model pull: Granite 4.0 Micro GGUF
Stale: Latest smoke run failed during artifact collection.