How the leader of leaders is scored
- 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. 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. 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. 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.
Every board and whether it counts
15 boards count toward the leader of leaders today.
| Board | Category | Metric | Published | Counts |
|---|---|---|---|---|
 LMArena Text Arenalmarena.ai leaderboard page data | Chat | Arena rating | Sep 13, 2026 | Yes |
 Scale SEAL MultiChallengescale.com leaderboard page | Chat | Score | Sep 24, 2026 | Yes |
 Artificial Analysis Intelligence Indexartificialanalysis.ai models page data | Reasoning | Index | Sep 24, 2026 | Yes |
| Humanity's Last Exam scale.com leaderboard page | Reasoning | Accuracy | Sep 24, 2026 | Yes |
 GPQA Diamond (Epoch AI runs)epoch.ai benchmark data (CC BY) | Reasoning | Accuracy | Sep 2, 2026 | Yes |
| Epoch Capabilities Index epoch.ai eci_scores.csv (CC BY) | Reasoning | ECI | Sep 24, 2026 | Yes |
 ARC-AGI-2arcprize.org leaderboard data files | Reasoning | Score | Sep 22, 2026 | Yes |
 LiveBenchlivebench.ai release tables | Reasoning | Global average | Jun 25, 2026 | Yes |
 MMLU-ProTIGER-Lab results.csv on Hugging Face | Reasoning | Accuracy | Stale since 2026-03-11 | Not fresh |
| LMArena WebDev Arena lmarena.ai leaderboard page data | Coding | Arena rating | Sep 23, 2026 | Yes |
| SWE-Bench Pro scale.com leaderboard page | Coding | Resolved | Sep 24, 2026 | Yes |
 SWE-bench Verifiedswe-bench.github.io leaderboards.json | Coding | Resolved | Stale since 2026-02-26 | Not fresh |
 Aider PolyglotAider polyglot_leaderboard.yml on GitHub | Coding | Pass rate | Stale since 2025-10-03 | Not fresh |
 Berkeley Function Calling Leaderboardgorilla.cs.berkeley.edu data_overall.csv | Agents and tools | Overall accuracy | Apr 12, 2026 | Yes |
| MCP Atlas scale.com leaderboard page | Agents and tools | Pass rate | Sep 23, 2026 | Yes |
| LMArena Search Arena lmarena.ai leaderboard page data | Agents and tools | Arena rating | Aug 24, 2026 | Yes |
| LMArena Vision Arena lmarena.ai leaderboard page data | Vision | Arena rating | Sep 13, 2026 | Yes |
| LMArena Document Arena lmarena.ai leaderboard page data | Vision | Arena rating | Sep 13, 2026 | Yes |
| LMArena Text-to-Image lmarena.ai leaderboard page data | Image | Arena rating | Sep 21, 2026 | Category only |
| LMArena Image Edit lmarena.ai leaderboard page data | Image | Arena rating | Sep 21, 2026 | Category only |
| LMArena Text-to-Video lmarena.ai leaderboard page data | Video | Arena rating | Sep 21, 2026 | Category only |
| LMArena Image-to-Video lmarena.ai leaderboard page data | Video | Arena rating | Sep 21, 2026 | Category only |
| Artificial Analysis Speech Arena artificialanalysis.ai speech arena page data | Speech | Arena rating | Sep 24, 2026 | Category only |
 Open ASR Leaderboardhf-audio results CSV on Hugging Face | Speech | Average WER (lower is better) | Sep 22, 2026 | Category only |
| MTEB Multilingual v2 MTEB leaderboard backend API | Embeddings | Mean task score | Sep 22, 2026 | Category only |
| Artificial Analysis Output Speed artificialanalysis.ai models page data | Speed | Output speed | Sep 24, 2026 | Category only |
| Artificial Analysis Price artificialanalysis.ai models page data | Price | Blended price (lower is better) | Sep 24, 2026 | Category only |
 OpenRouter weekly usageopenrouter.ai rankings page data | Usage | Weekly tokens | Sep 23, 2026 | Category only |
| Hugging Face Open LLM Leaderboard retired, not captured | Reasoning | Average | Retired | Category only |
What this is not
A board is one team’s method on one day. Leaders Board 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.
Scores as published by each board on the capture date. Model names and logos belong to their owners; logos via logo.dev.