Advanced CLI tool that scans your hardware and tells you exactly which LLM or sLLM models you can run locally, with full Ollama integration.
2.8k
Stars
188
Forks
5
Open issues
4
Contributors
AI Analysis
LLM Checker is a specialized CLI tool that analyzes your hardware and recommends which LLM models you can run locally, with integration to Ollama and a curated registry of 33k+ model artifacts. It serves ML engineers and developers exploring local inference optimization; it is not a general-purpose tool and does not help users who need cloud-based model serving or have no interest in running models locally.
Inferred from signals mentioned in the README (tests, CI, type safety) — not a review of the actual code.
AI's overall editorial judgment — not an average of the bars above, can weigh other factors too.
JavaScript CLI that matches local LLM models to your hardware via multi-source catalog and Ollama integration
LLM Checker is a Node.js CLI tool designed to help users identify which large language models can run on their specific hardware. It analyzes system capabilities (GPU type, VRAM, CPU) and scores 33k+ model variants from Hugging Face, Ollama, and GPT4All across four dimensions (Quality, Speed, Fit, Context), then recommends compatible models with download/run commands. Built primarily for developers and AI enthusiasts running models locally via Ollama. Adoption appears concentrated among early adopters; real-world production usage not publicly documented.
Project created July 2025, approximately 11 months old at analysis date. Entered open source ecosystem as a specialized tool addressing the practical friction point of model selection on heterogeneous local hardware. No evidence of commercial origin or major organizational backing mentioned in README.
Gained 51 stars in the week preceding analysis (2026-06-22 to 2026-06-29), indicating recent momentum. Total 2,738 stars and 180 forks suggest niche but engaged audience. Growth pattern appears consistent with early-stage specialized developer tool discovery, but momentum is modest relative to larger ecosystem alternatives (vllm: 84.7k stars, llmfit: 28.7k stars). npm package distribution suggests awareness-building phase.
Adoption not verified. README contains no case studies, testimonials, company names, or quantified user counts. npm download badges referenced but no actual monthly download figures provided in excerpt. Discord community mentioned but size unknown. Project appears in early adoption phase with enthusiast audience; enterprise or production deployment not evident from public signals.
Based on README: pure JavaScript implementation with zero native dependencies, runs on Node.js 16+. Includes prebuilt SQLite catalog of ~200 Ollama models, optional sql.js for database operations, and integration with Ollama daemon for model discovery and execution. Appears to ship a deterministic scoring engine that weights four dimensions. Multi-source registry management suggests layered data structure. Likely uses subprocess or IPC to invoke Ollama CLI.
Not documented in README. No testing framework, test command, or coverage metrics mentioned.
Last push 2026-06-22 (7 days before analysis date) indicates active maintenance. README references recent features (Claude MCP integration, calibration system, ai-run metrics). Changelog documented in repo. Discord community referenced. Maintenance appears current and responsive, though commit frequency cannot be verified from metadata alone.
ADOPT IF: you run models locally via Ollama and need quick hardware-aware recommendations across diverse GPU types (Apple Silicon, NVIDIA, AMD, Intel Arc); you prefer JavaScript tooling in your local workflow; your hardware is not exotic or highly constrained. AVOID IF: you need production deployment automation, run inference at scale, or require cutting-edge optimization (vllm is better); your workflow is already mature with a chosen model; you need proven, long-stable tooling (whichllm or commercial solutions have longer track records). MONITOR IF: you want a lightweight discovery layer and can tolerate occasional mismatches between estimated and actual model compatibility; you value active development and early feature adoption.
Independent dimensions
Mainstream potential
3/10
Technical importance
6/10
Adoption evidence
2/10
- Adoption not verified at production scale — unclear how often recommendations match real-world success on diverse hardware.
- Narrow specialization to Ollama ecosystem means less relevance if users migrate to other inference runtimes (vLLM, llama.cpp directly, etc.).
- Test coverage not documented; code quality and regression handling cannot be assessed from metadata.
- Small maintainer team (inferred from single-author GitHub profile) may face scaling or burnout risk if adoption grows significantly.
- Hardware calibration data (bytes-per-parameter formula) not independently verified; incorrect estimates could lead to out-of-memory failures if used in production automation.
LLM Checker likely remains a specialized discovery tool within the Ollama ecosystem, maintaining modest but engaged adoption among local inference enthusiasts. Unlikely to achieve mainstream dominance due to narrow scope and strong incumbent positioning (vllm ecosystem, commercial offerings). May consolidate into broader inference frameworks or stabilize as a reference tool cited in tutorials.
Newsletter
Get analyses like this every Monday
Free weekly digest of the most interesting open-source discoveries.
Languages
Information
- Language
- JavaScript
- License
- NOASSERTION
- Last updated
- 4d ago
- Created
- 12mo ago
- Analyzed with
- anthropic/claude-haiku-4-5
Stars over time
Contributors over time
Top 100 contributors only — repos with more will plateau at 100.
Open issues
Top contributors
Recent releases
Similar repos
Andyyyy64/whichllm
whichllm is a CLI tool that automatically detects your hardware (GPU, CPU, RAM)...
| Repository | Stars | Week Δ | Language | Score | Updated |
|---|---|---|---|---|---|
|
signerless/llm-checker
this
|
2.8k | +34 | JavaScript | 8/10 | 4d ago |
 AlexsJones/llmfit
|
29.3k | — | Rust | 8/10 | -4 min ago |
|
Andyyyy64/whichllm
|
5.7k | — | Python | 8/10 | 2d ago |
 jamesob/local-llm
|
1.3k | — | Shell | 8/10 | 2d ago |
 vllm-project/guidellm
|
1.4k | — | Python | 7/10 | 10h ago |
 ollama/ollama
|
175.9k | — | Go | 9/10 | 12h ago |
llmfit focuses on broader LLM workflow support and model-fit evaluation. LLM Checker specializes narrowly in hardware-aware model selection. README explicitly positions them as complementary rather than competing, suggesting different use cases.
vLLM is a production inference optimization library. LLM Checker is a model discovery and recommendation CLI. Different problems: vLLM accelerates serving, LLM Checker helps users choose what to serve. No direct overlap.
Functionally similar (hardware-aware model recommendation). whichllm has 2x the stars and predates LLM Checker. LLM Checker differentiates with Ollama integration, multi-source registry (33k artifacts vs. undocumented in whichllm), and JavaScript accessibility.
mlc-llm is a cross-platform model compilation and serving framework. LLM Checker is a discovery tool. Complementary rather than competitive; mlc-llm users may use LLM Checker for initial model selection.
guidellm is a benchmarking/guidance tool for vLLM. LLM Checker is a recommendation CLI. No direct overlap; different problem domains.
