# Andyyyy64/whichllm

Find the local LLM that actually runs and performs best on your hardware. Ranked by real, recency-aware benchmarks, not parameter count. One command, run it instantly.

Repository: https://github.com/Andyyyy64/whichllm
Canonical: https://ross.abutalabs.com/products/whichllm
Language: Python
License: MIT
License Family: permissive
Topics: localllm
Last push: 2026-08-14T06:49:58+00:00

## Health v2 (maintenance only)
Score: 81/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 97, release rhythm 98, longevity 13
- inputs: {"age_days": 182, "days_push": 19, "days_rel": 19, "gap_med": 2.5, "n_releases_24m": 17}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 6488, forks 354 (observed 2026-08-28T04:09:43.770443+00:00)

## What it is
whichllm is a Python CLI tool that auto-detects your GPU, CPU, and RAM, then ranks local LLMs from HuggingFace that will actually run and perform well on your hardware. It uses recency-aware benchmarks rather than parameter count and supports simulating GPUs before purchase.

## Use cases
- find the best local LLM that runs on my GPU
- recommend a model that fits my VRAM
- check which LLM my hardware can run before buying a GPU
- compare models for RTX 4090 vs RTX 5090
- find what GPU I need to run llama 3 70b
- get a list of fast local models for my machine

## When to choose
- you want a one-command recommendation for local LLMs matched to your hardware
- you're deciding which GPU to buy and want to simulate its model capacity
- you want recency-aware rankings instead of guessing by parameter count

## When to avoid
- you need to actually serve or deploy models at scale rather than pick one
- you want cloud/API model recommendations
- you need fine-grained benchmarking of inference throughput on your own workloads

## Facets
- artifact type: cli-tool
- maturity: active
- function: llm-inference, cli, developer-tools
- domain: large-language-models, machine-learning, developer-tools
- platform: cli, python, cross-platform, windows
- tags: local-llm, hardware-recommendation, gpu, vram, huggingface, model-selection, benchmarking, command-line, macos, linux

## Member repositories
- Andyyyy64/whichllm (main) score 81

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:43.770443+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:44:31.961079+00:00, confidence not recorded.
  - readme: https://github.com/Andyyyy64/whichllm (fetched 2026-08-28T04:09:43.770443+00:00, sha 611eaab75254)
  - registry_pypi: https://pypi.org/pypi/whichllm/json (fetched 2026-08-29T08:41:04.223982+00:00, sha 881e71b379ff)
- Data as of 2026-08-30T08:39:29.467469+00:00.