Digital Work Index
Gemini 4 Argon 28.3 leads since October 2026 Models are scored on eight benchmarks in coding, reasoning, writing and enterprise work. Weights come from the wages of the 29 digital occupations the benchmarks map to.
Weights use BLS employment and May 2025 wages for 237 digital occupations. The board had 16 models at launch.
- Made by
- Snorkel AI
- Released
- October 2026
- Kind
- Composite work indices
- Level
- Expert
- Scored
- Graded by a model
- Answers
- Agent task
- Tasks
- Kept private
- Strongest model tested
- GPT-6 Astra #3 on the capability index
- Last model added
- October 2026
- Scores refreshed
- November 12th, 2026
Gemini 4 Argon leads with 28.3
Best score per model from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best.
- 1 Gemini 4 Argon 28.3
- 2
Muse Wonder 26.4 - 3 Claude Fable 5.1 23.6
- 4 Claude Opus 5.5 23.6
- 5 GPT-6 Astra 23.0
- 6 Claude Opus 5 21.8
- 7 GPT-6.1 Sol 21.2
- 8 Grok 4.6 20.5
- 9 Grok 4.7 19.3
- 10 Gemini 3.8 Flash 16.9
- 11 Kimi K3 16.8
- 12 GLM-5.3 16.4
- 13 Muse Spark 1.3 16.3
- 14
Qwen 3.8 Max 11.8 - 15
DeepSeek-V4-Pro 11.1 - 16
Nemotron 3 Ultra 6.95
Source: Official Digital Work Index leaderboard
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Gemini 4 Argon | 28.3 | Sep 2026 | Official leaderboard | |
| 2 | Muse Wonder | 26.4 | Oct 2026 | Official leaderboard | |
| 3 | Claude Fable 5.1 | 23.6 | Sep 2026 | Official leaderboard | |
| 4 | Claude Opus 5.5 | 23.6 | Sep 2026 | Official leaderboard | |
| 5 | GPT-6 Astra | 23.0 | Sep 2026 | Official leaderboard | |
| 6 | Claude Opus 5 | 21.8 | Jul 2026 | Official leaderboard | |
| 7 | GPT-6.1 Sol | 21.2 | Sep 2026 | Official leaderboard | |
| 8 | Grok 4.6 | 20.5 | Aug 2026 | Official leaderboard | |
| 9 | Grok 4.7 | 19.3 | Sep 2026 | Official leaderboard | |
| 10 | Gemini 3.8 Flash | 16.9 | Sep 2026 | Official leaderboard | |
| 11 | Kimi K3 | 16.8 | Jul 2026 | Official leaderboard | |
| 12 | GLM-5.3 | 16.4 | Aug 2026 | Official leaderboard | |
| 13 | Muse Spark 1.3 | 16.3 | Sep 2026 | Official leaderboard | |
| 14 | Qwen 3.8 Max | 11.8 | Aug 2026 | Official leaderboard | |
| 15 | DeepSeek-V4-Pro | 11.1 | Apr 2026 | Official leaderboard | |
| 16 | Nemotron 3 Ultra | 6.95 | Jun 2026 | Official leaderboard |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official Digital Work Index leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official Digital Work Index leaderboard
No lab model card reports it yet
We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.
More in Economic work
- JobBench In use 130 tasks from 35 occupations workers most want delegated. Agent works through messy reference files, graded by fact-anchored binary rubrics. GLM-5.3 61.4%
- GDPval In use Real work tasks from 44 occupations with reference files. Experts blindly compare model deliverables with expert work. Metric is win-or-tie rate.
Seed2.1 Pro 87.9% - GDPval-AA v2.1 In use Agents with shell and web access complete 220 GDPval gold tasks from 44 occupations. Blind pairwise comparisons give Elo ratings. Claude Opus 5.5 1,846
- AA-Briefcase v1.1 In use Version 1.1 of Artificial Analysis's multi-hour office work evaluation, with Elo ratings fitted by Crowd-BT. Claude Opus 5.5 1,822
- CoWorkBench In use Agent completes long-horizon productivity tasks across computer science, finance, law and medicine. Qwen in-house scoring (method unpublished). Claude Fable 5 75.9%
- OfficeQA In use Grounded reasoning over a large corpus of U.S. Treasury Bulletin documents: locate tables and compute precise numeric answers. Claude Opus 4.7 86.3%