leaderboard 开源项目推荐（1,823 个）

Cloud-CV /EvalAI

：cloud：：rocket：：bar_chart：：chart_with_updates_trend：评估人工智能的最新水平

tatsu-lab /alpaca_eval

An automatic evaluator for instruction-following language models. Human-validated, high-quality, cheap, and fast.

THUDM /CogDL

CogDL: A Comprehensive Library for Graph Deep Learning (WWW 2023)

suyoumo /ClawProBench

ClawProBench is a live-first benchmark harness for evaluating LLM agents in the OpenClaw runtime with deterministic grading and repeated-trial reliability.

cjmellor /level-up

Level-Up 是一个 Laravel 软件包，将游戏化引入你的应用中。用户通过互动获得经验值（XP）和等级，同时解锁成就。它通过动态排行榜功能促进参与、竞争和乐趣。可根据你的具体需求进行定制

onejune2018 /Awesome-LLM-Eval

Awesome-LLM-Eval: a curated list of tools, datasets/benchmark, demos, leaderboard, papers, docs and models, mainly for Evaluation on LLMs. 一个由工具、基准/数据、演示、排行榜和大模型等组成的精选列表，主要面向基础大模型评测，旨在探求生成式AI的技术边界.

rajpurkar /SQuAD-explorer

Visually Explore the Stanford Question Answering Dataset

agent-next /polymarket-paper-trader

Polymarket 的纸币交易模拟器——专为人工智能代理打造。MCP服务器、实时订单簿、策略回测。安装：npx clawhub 安装 polymarket-paper-trader

waico /SKAB

SKAB - Skoltech Anomaly Benchmark. Time-series data for evaluating Anomaly Detection algorithms.

SAILResearch /awesome-ai-leaderboard

A curated list of awesome leaderboard-oriented resources for AI domain

BichengWang /RL-pricing

A reproducibility lab for trading AI: published RL and LLM trading claims re-tested under a pre-registered protocol, with a causal benchmark and a live track.

fstandhartinger /jevbench

JevBench by Benchmark Heaven: benchmark and leaderboard for AI decision models, measuring typed decision accuracy, calibration, latency and cost.

gpu-mode /reference-kernels

Official Problem Sets / Reference Kernels for the GPU MODE Leaderboard!

xcpcio /xcpcio

The ICPC Series Competition Leaderboard Visualization Engine

LudwigStumpp /llm-leaderboard

A joint community effort to create one central leaderboard for LLMs.

medmcqa /medmcqa

A large-scale (194k), Multiple-Choice Question Answering (MCQA) dataset designed to address realworld medical entrance exam questions.

FindAny /ghfind

Discover the best developers — and become one. Drop a GitHub handle for a 0–100 value & trust score in 30s: see your gaps, discover top devs, get found. Exposes PR farmers, AI bots & fork-hoarders. Deterministic scoring, self-hostable.

liyupi /ai-model-world

AI 大模型世界，把 556 个大模型拟人化成像素小人的可视化站点。进来就能看到此刻谁最聪明、谁最会写代码、谁最便宜、谁刚发布，往下是国内与国外分区的厂商广场、完整的发布时间线和多维排行榜。搜索认模型名、厂商和能力，输入「多模态」会直接列出全部多模态模型。数据取自 Epoch AI、models.dev、LiveBench 与 Hugging Face，每小时自动同步，所有文案由真实数据生成，不调用任何 LLM。Next.js 静态导出，零后端。

AI4LIFE-GROUP /OpenXAI

OpenXAI : Towards a Transparent Evaluation of Model Explanations

lechmazur /confabulations

Hallucinations (Confabulations) Document-Based Benchmark for RAG. Includes human-verified questions and answers.