benchmarks.site

Public leaderboard · Aggregated leaderboards

Scale SEAL Leaderboards

Scale AI tests AI models on private, expert-written questions that models cannot have memorised, across areas like coding, reasoning and safety. One representative board (Humanity's Last Exam, Diamond) is shown.

Leaderboard

Top 3 right now

  1. 1stclaude-fable-5-272.4
  2. 2ndgpt-6-astra60.6
  3. 3rdclaude-opus-5-555

What it measures

Collection of expert-built, mostly private evaluations (such as SWE-Bench Pro, MCP Atlas, MultiChallenge, Remote Labor Index, HLE) run by Scale.

How to read the numbers

Share of expert-level questions answered correctly in Humanity's Last Exam (Diamond) (%) — higher is better.

Results

6 rows · higher is better (Score (%))CSV
1claude-fable-5-2anthropic72.42.8
2gpt-6-astraopenai60.63
3claude-opus-5-5anthropic553.1
4gemini-4-argongoogle53.23.07
5claude-fable-5-1anthropic51.33.1
6claude-opus-5anthropic38.63

Note: SEAL hosts ~30 leaderboards; only HLE (Diamond) shown, non-deprecated entries. as_of is the board's updatedAt field; newest entries dated Nov 2026. Some entries carry a contamination warning on the site. Data read from page-embedded JSON.

Source: Scale AI (SEAL) · labs.scale.com · results as of 2026-09-23 · retrieved 2026-11-17

Markdown · More Aggregated leaderboards