# llm2014/llm_benchmark

Repository: https://github.com/llm2014/llm_benchmark
Canonical: https://ross.abutalabs.com/products/llm_benchmark
License Family: other
Last push: 2026-08-26T16:28:45+00:00

## Health v2 (maintenance only)
Score: 65/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 35, longevity 40
- inputs: {"age_days": 572, "days_push": 7, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1567, forks 22 (observed 2026-08-28T04:05:05.020097+00:00)

## What it is
A personal, long-running benchmark that tracks large language model performance on logic, math, programming, and intuition tasks using a private, rolling question bank of ~28 questions. Results are published as a monthly leaderboard with scoring based on multi-point rubrics.

## Use cases
- compare reasoning ability of different LLMs
- track how LLMs improve over time
- find which model is best at math and logic puzzles
- view a leaderboard of large language model scores
- evaluate models on programming and deduction tasks

## When to choose
- you want an independent, long-term view of LLM reasoning trends
- you need a leaderboard covering logic, math, and coding ability
- you want to see how specific models evolve month over month

## When to avoid
- you need a comprehensive or authoritative academic benchmark
- you require publicly available test questions to run yourself
- you need domain-specific evaluation outside reasoning tasks

## Facets
- artifact type: dataset
- maturity: active
- function: benchmarking, llm-inference
- domain: large-language-models, artificial-intelligence, machine-learning
- platform: python
- tags: llm-evaluation, leaderboard, private-question-bank, reasoning-benchmark, monthly-updates, chinese-language, evaluation, web-server

## Member repositories
- llm2014/llm_benchmark (main) score 65

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:05.020097+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 2026-08-30T03:59:01.011433+00:00, confidence not recorded.
  - readme: https://github.com/llm2014/llm_benchmark (fetched 2026-08-28T04:05:05.020097+00:00, sha b4bd052f13c4)
- Data as of 2026-08-30T08:39:29.467469+00:00.