Snorkel Digital Work Index Coding
Indexes
Published coding panel of the Snorkel Digital Work Index. Diagnostic only because the constituent coding benchmarks are represented independently.
Top models (higher is better)
| Model | Score |
|---|---|
| GPT-6 Astra | 28.8 |
| Claude Opus 5.5 | 27.7 |
| Claude Fable 5.1 | 26.2 |
| Claude Opus 5 | 25.6 |
| Grok 4.6 | 18.7 |
| Gemini 3.8 Flash | 18.5 |
| Grok 4.7 | 18.1 |
| Kimi K3 | 14.3 |
| GLM-5.3 | 13.4 |
| Muse Spark 1.3 | 12.8 |