Every AI leaderboard

Leaderboards28Models ranked986

28 leaderboards · 986 models ranked

How the leader of leaders is scored

Every step, so you can check it or disagree with it.

  1. 1. Capture

    Every board is fetched daily at 04:30 UTC from its own page, data file or API and parsed without a model. If a parser breaks, the model reads that board’s page instead and those rows carry an Unverified badge.

  2. 2. One model, one entry

    Variants collapse to the model: effort levels (high, xhigh, max), thinking modes, dated snapshots and agent harnesses. The best variant on a board stands for the model there. The exact name each board published is kept next to every entry.

  3. 3. Which boards count

    Language-model quality boards (chat, reasoning, coding, agents and tools, vision) that published in the last 180 days. Image, video, speech, embeddings, speed, price and usage get their own category leaders.

  4. 4. Points

    On each counted board: #1 earns 10 points, #2 earns 9, down to 1 point for #10. Podium score = points earned / (10 × boards counted) × 100. Ties break on #1 finishes, then the median rank across boards that list the model.

101
92
83
74
65
56
47
38
29
110

Which boards count: 14 of 29 today

BoardCounts
LMArena Text ArenaLMArena leaderboard dataset on Hugging Face Yes
Scale SEAL MultiChallengelabs.scale.com leaderboard page Yes
Artificial Analysis Intelligence Indexartificialanalysis.ai models page data Yes
Humanity's Last Examlabs.scale.com leaderboard page Yes
GPQA Diamond (Epoch AI runs)epoch.ai benchmark data (CC BY) Yes
Epoch Capabilities Indexepoch.ai eci_scores.csv (CC BY) Yes
ARC-AGI-2arcprize.org leaderboard data files Yes
LiveBenchlivebench.ai release tables Yes
MMLU-ProTIGER-Lab results.csv on Hugging FaceNot fresh
LMArena WebDev ArenaLMArena leaderboard dataset on Hugging Face Yes
SWE-Bench Prolabs.scale.com leaderboard page Yes
SWE-bench Verifiedswe-bench.github.io leaderboards.jsonNot fresh
Aider PolyglotAider polyglot_leaderboard.yml on GitHubNot fresh
Berkeley Function Calling Leaderboardgorilla.cs.berkeley.edu data_overall.csvNot fresh
MCP Atlaslabs.scale.com leaderboard page Yes
LMArena Search ArenaLMArena leaderboard dataset on Hugging Face Yes
LMArena Vision ArenaLMArena leaderboard dataset on Hugging Face Yes
LMArena Document ArenaLMArena leaderboard dataset on Hugging Face Yes
LMArena Text-to-ImageLMArena leaderboard dataset on Hugging FaceCategory only
LMArena Image EditLMArena leaderboard dataset on Hugging FaceCategory only
LMArena Text-to-VideoLMArena leaderboard dataset on Hugging FaceCategory only
LMArena Image-to-VideoLMArena leaderboard dataset on Hugging FaceCategory only
Artificial Analysis Speech Arenaartificialanalysis.ai speech arena page dataCategory only
Open ASR Leaderboardhf-audio results CSV on Hugging FaceCategory only
MTEB Multilingual v2MTEB leaderboard backend APICategory only
Artificial Analysis Output Speedartificialanalysis.ai models page dataCategory only
Artificial Analysis Priceartificialanalysis.ai models page dataCategory only
OpenRouter weekly usageopenrouter.ai rankings page dataCategory only
Hugging Face Open LLM Leaderboardretired, not capturedCategory only

What this is not

A board is one team’s method on one day. This site does not run evaluations and does not adjust anyone’s numbers; it shows where the boards agree. Vendors: OpenAI, Anthropic, Google and others appear by the name each board uses, linked to their own pages.

All boards

How the list of boards grows

Each run looks for leaderboards not read yet: the official leaderboard of every benchmark in the catalog, Hugging Face Spaces that are leaderboards (100+ likes, updated in the last year) and a weekly search. A found board is checked (it must answer) and listed here until a reader for it is written; then it joins the table above. 143 models were added in the last 30 days. 60 found so far, 2 more to check.