Remote Labor Index
ReportedNo description yet: we only describe an eval set when we can do it correctly.
- Kind
- Published by a third party's leaderboard, collected by Epoch AI. The leaderboard is named and linked.
- Score shown
- Epoch's column "Score", as a percentage, higher is better; random guessing scores about 0.0%. The source gives no error bars.
- Results
- 20 in all: 1 for open-weight models, 19 for closed ones. 1 linked to a model page here.
Best open-weight result: deepseek-ai/DeepSeek-V4.1-Flash, 9.6%, number 8 of 20 results.
| # | Model | Weights | Setting | Score | Date | Source |
|---|---|---|---|---|---|---|
| 1 | claude-fable-5-2_unknownAnthropic | closed | effort unknown | 41.3% | not dated | via Epoch AI |
| 2 | claude-opus-5-5_unknownAnthropic | closed | effort unknown | 21.3% | not dated | via Epoch AI |
| 3 | gpt-6-astra_unknownOpenAI | closed | effort unknown | 20.8% | not dated | via Epoch AI |
| 4 | claude-fable-5-1_unknownAnthropic | closed | effort unknown | 17.9% | not dated | via Epoch AI |
| 5 | gemini-4-argon_unknownGoogle DeepMind | closed | effort unknown | 17.1% | not dated | via Epoch AI |
| 6 | claude-fable-5Anthropic | closed | default | 16.1% | not dated | via Epoch AI |
| 7 | claude-fable-5_unknownAnthropic | closed | effort unknown | 15.8% | not dated | via Epoch AI |
| 8 | deepseek-ai/DeepSeek-V4.1-FlashEpoch: deepseek-v4.1-flash_unknown, DeepSeek | open | effort unknown | 9.6% | not dated | via Epoch AI |
| 9 | claude-opus-4-8_unknownAnthropic | closed | effort unknown | 8.3% | not dated | via Epoch AI |
| 10 | gpt-5.5_unknownOpenAI | closed | effort unknown | 6.3% | not dated | via Epoch AI |
| 11 | gemini-3.8-flash_unknownGoogle DeepMind | closed | effort unknown | 5.8% | not dated | via Epoch AI |
| 12 | gemini-3.7-flash_unknownGoogle DeepMind | closed | effort unknown | 5.0% | not dated | via Epoch AI |
| 13 | claude-opus-4-6_unknownAnthropic | closed | effort unknown | 4.2% | not dated | via Epoch AI |
| 14 | claude-opus-4-5-20251101_unknownAnthropic | closed | effort unknown | 3.8% | not dated | via Epoch AI |
| 15 | gpt-5.2-2025-12-11_mediumOpenAI | closed | effort medium | 2.5% | not dated | via Epoch AI |
| 16 | claude-sonnet-4-5-20250929_unknownAnthropic | closed | effort unknown | 2.1% | not dated | via Epoch AI |
| 17 | gpt-5.2-2025-12-11_unknownOpenAI | closed | effort unknown | 2.1% | not dated | via Epoch AI |
| 18 | gpt-5-2025-08-07_unknownOpenAI | closed | effort unknown | 1.7% | not dated | via Epoch AI |
| 19 | gemini-3-pro-previewGoogle DeepMind | closed | default | 1.3% | not dated | via Epoch AI |
| 20 | gemini-2.5-pro-preview-06-05Google DeepMind | closed | default | 0.8% | not dated | via Epoch AI |
Open-weight rows are shaded. Dates are when the result was run or evaluated, where the data says. Data from Epoch AI (CC-BY 4.0), fetched 19 Nov 2026: Epoch AI, ‘Capabilities & benchmarking’. Published online at epoch.ai. Retrieved from ‘https://epoch.ai/benchmarks’ [online resource]. Epoch AI. Method