"LLM-Checker: The CLI Tool That Tells You Which Local LLMs Actually Fit Your Hardware"

𝐋𝐋𝐌-𝐂𝐡𝐞𝐜𝐤𝐞𝐫 - 𝐋𝐋𝐌 𝐑𝐞𝐜𝐨𝐦𝐦𝐞𝐧𝐝𝐚𝐭𝐢𝐨𝐧 𝐓𝐨𝐨𝐥 𝐟𝐨𝐫 𝐘𝐨𝐮𝐫 𝐋𝐨𝐜𝐚𝐥 𝐒𝐲𝐬𝐭𝐞𝐦

Are you struggling to figure out which LLMs you can run locally on your system? LLM Checker CLI tool comes to the rescue.

𝐋𝐋𝐌 𝐂𝐡𝐞𝐜𝐤𝐞𝐫 𝐢𝐬 𝐚𝐧 𝐚𝐝𝐯𝐚𝐧𝐜𝐞𝐝 𝐂𝐋𝐈 𝐭𝐨𝐨𝐥 𝐭𝐡𝐚𝐭

• scans your local system hardware (CPU, RAM, GPU) and
• recommends which LLMs you can run locally with Ollama

If you build or deploy Local GenAI systems, LLM Checker is a must in your toolkit.

𝐅𝐞𝐚𝐭𝐮𝐫𝐞𝐬

• 35+ Curated Models: Hand-picked catalog covering all major families and sizes (1B-32B)
• 4D Scoring Engine: Quality, Speed, Fit, Context — weighted by use case
• Multi-GPU Hardware Detection: Apple Silicon, NVIDIA CUDA, AMD ROCm, Intel Arc, CPU
• Calibrated Memory Estimation: Bytes-per-parameter formula validated against real Ollama sizes
• Zero Native Dependencies: Pure JavaScript — works on any Node.js 16+ system

llm-checker is a real, actively maintained open-source npm package on GitHub (Pavelevich/llm-checker, also mirrored as signerless/llm-checker), and the post's claims match the tool's actual documented behavior closely:

• Hardware scanning + Ollama recommendations — confirmed. It's an advanced CLI tool that analyzes your hardware and intelligently recommends the optimal Ollama LLM models for your system with automatic installation detection. Socket
• 4D Scoring (Quality, Speed, Fit, Context) — confirmed exactly. It analyzes your system, scores every compatible model across four dimensions (Quality, Speed, Fit, Context), and delivers actionable recommendations in seconds. npm
• Node.js 16+, cross-platform — confirmed. Engine requirement is Node.js >=16.0.0, and it supports android, darwin, linux, and win32. GitHubGitHub
• Multi-GPU/hardware detection (Apple Silicon, CUDA, etc.) — consistent with real tiering shown in docs (RTX 5090/H100, RTX 4090/A100, Apple Silicon tiers are used to classify hardware) Socket
• Calibrated memory estimation — matches real described functionality: "hardware-calibrated memory estimation" is an actual named feature npm

One discrepancy:

• "35+ Curated Models" — this figure is outdated/understated for current versions. More recent releases show the catalog has grown substantially: one source notes 200+ dynamic models with 35+ curated as fallback, and another cites 33k+ exact artifacts across a multi-source registry (Hugging Face + Ollama + GPT4All). So "35+" was accurate for an earlier version but likely undersells the current tool's scale — not fake, just stale