SEAL - Remote Labor Index (RLI): leaderboard
240 end-to-end freelance projects from Upwork across 23 professional domains with diverse deliverable types (documents, audio, video, 3D models), measuring practical work automation.
Metric: Score. Source: scale.com. Saturation forecast: Around November 2027. 20 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5.5 | 21.25 | #3 |
| 2 | GPT-6 Astra | 20.83 | #6 |
| 3 | Claude Fable 5.1 | 17.92 | #9 |
| 4 | Claude Fable 5 | 15.83 | #16 |
| 5 | Claude Opus 4.8 | 8.33 | #37 |
| 6 | GPT-5.5 | 6.25 | #31 |
| 7 | Gemini 3.8 Flash | 5.83 | #21 |
| 8 | Gemini 3.7 Flash | 5 | #23 |
| 9 | Claude Opus 4.6 | 4.17 | #63 |
| 10 | Claude Opus 4.5 (20251101) (Thinking) | 3.75 | #74 (Claude Opus 4.5 (20251101)) |
| 11 | GPT-5.2 (2025-12-11) (Medium) | 2.5 | |
| 12 | Claude Sonnet 4.5 | 2.08 | #142 |
| 13 | GPT-5 | 1.67 | #98 |
| 14 | Gemini 3 Pro (Preview) | 1.25 | #66 |
| 15 | Gemini 2.5 Pro (Preview 06-05) | 0.83 | #106 |
No result here: #7 GPT-6.1 Sol, #8 Gemini 4 Argon, #10 Muse Wonder Full.
Interactive version: theaggregate.ai/benchmark?slug=seal-remote-labor-index-rli · How It Works · Data refreshed daily, snapshot 2026-10-20.