What's changed\n- Security hardening for banner loading (removed executable parsing path, JSON-only external banners).\n- Dependency security updates and lockfile refresh to address audit findings.\n- Release prep for npm: package name/version normalized to \
+--------------------------------------------------------------------------------------------------------------------------------------------------+
| |
| [ INTELLIGENT OLLAMA MODEL SELECTOR ] |
|---|
| █████ █████ ██████ ██████ |
| ▒▒███ ▒▒███ ▒▒██████ ██████ |
| ▒███ ▒███ ▒███▒█████▒███ |
| ▒███ ▒███ ▒███▒▒███ ▒███ |
| ▒███ ▒███ ▒███ ▒▒▒ ▒███ |
| ▒███ █ ▒███ █ ▒███ ▒███ |
| ███████████ ███████████ █████ █████ |
| ▒▒▒▒▒▒▒▒▒▒▒ ▒▒▒▒▒▒▒▒▒▒▒ ▒▒▒▒▒ ▒▒▒▒▒ |
| █████████ █████ █████ ██████████ █████████ █████ ████ ██████████ ███████████ |
| ███▒▒▒▒▒███▒▒███ ▒▒███ ▒▒███▒▒▒▒▒█ ███▒▒▒▒▒███▒▒███ ███▒ ▒▒███▒▒▒▒▒█▒▒███▒▒▒▒▒███ |
| ███ ▒▒▒ ▒███ ▒███ ▒███ █ ▒ ███ ▒▒▒ ▒███ ███ ▒███ █ ▒ ▒███ ▒███ |
| ▒███ ▒███████████ ▒██████ ▒███ ▒███████ ▒██████ ▒██████████ |
| ▒███ ▒███▒▒▒▒▒███ ▒███▒▒█ ▒███ ▒███▒▒███ ▒███▒▒█ ▒███▒▒▒▒▒███ |
| ▒▒███ ███ ▒███ ▒███ ▒███ ▒ █▒▒███ ███ ▒███ ▒▒███ ▒███ ▒ █ ▒███ ▒███ |
| ▒▒█████████ █████ █████ ██████████ ▒▒█████████ █████ ▒▒████ ██████████ █████ █████ |
| ▒▒▒▒▒▒▒▒▒ ▒▒▒▒▒ ▒▒▒▒▒ ▒▒▒▒▒▒▒▒▒▒ ▒▒▒▒▒▒▒▒▒ ▒▒▒▒▒ ▒▒▒▒ ▒▒▒▒▒▒▒▒▒▒ ▒▒▒▒▒ ▒▒▒▒▒ |
| AI-powered CLI for hardware-aware local LLM recommendations |
| Deterministic scoring across 200+ dynamic models (35+ curated fallback) |
| [200+ DYNAMIC MODELS] [35+ FALLBACK] [4D SCORING] [MCP SERVER] |
| ---------------------------------------------------------------------------------------------- |
| Install: npm install -g llm-checker |
| Run: llm-checker recommend |
| github.com/Pavelevich/llm-checker |
| +--------------------------------------------------------------------------------------------------------------------------------------------------+ |
Usage: llm-checker [options] [command]
Check which LLM models your computer can run
Options:
-V, --version output the version number
-h, --help display help for command
Commands:
help [command] Show all commands and how to use them
mcp-setup [options] Show or apply Claude MCP setup for llm-checker
policy Manage enterprise policy files (policy.yaml)
audit Run policy audits and export compliance reports
calibrate [options] Generate calibration contract artifacts from a
JSONL prompt suite
check [options] Analyze your system and show compatible LLM models
ollama [options] Manage Ollama integration with hardware
compatibility
installed [options] Show ranking of installed Ollama models by
compatibility and use-case
ollama-plan [options] Plan safe Ollama runtime settings for selected
local models
recommend [options] Get intelligent model recommendations for your
hardware
list-models [options] List all models from Ollama database
ai-check [options] AI-powered model evaluation with meta-analysis
ai-run [options] AI-powered model selection and execution
demo Demo of the enhanced verbose progress with
progress bars
sync [options] Sync the model database from Ollama registry
(scrapes all models)
search [options] Search models in the database with intelligent
scoring
smart-recommend [options] Get intelligent model recommendations using the
new scoring engine
hw-detect [options] Detect and display detailed hardware capabilities, publish config cleanup, and improved npm packaging excludes.\n\n## Verification\n- Test suite passing (23/23).\n- npm audit reports 0 vulnerabilities.\n- npm pack --dry-run validated clean tarball contents.