# Language (Liquid report)

> Liquid AI reports 56.4 skill points for d1-3B and 12.9 for d1-omni-600M on Language with the official Decision Index 0.2.1 scorer.

Canonical page: https://benchlm.ai/benchmarks/decision-index-language-liquid

- Category: [Decision Models](/decision-models)
- Last updated: October 7, 2026 provider report

## About Language (Liquid report)

- Year: 2026
- Format: Chance-adjusted skill points
- Difficulty: Varies by task and supplied answer options
- Paper: [Open d1: Edge decision models for text, vision, and audio](https://www.liquid.ai/blog/d1-open)

Liquid AI evaluated these checkpoints with the official Decision Index 0.2.1 scorer on the public split. These are provider-reported runs, not public leaderboard submissions. The index and five areas are skill points, not percent accuracy. They remain separate from Decision Index 0.3, JevBench, and general model rankings; no private vision-split result is reported.

Language (Liquid report) is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (2 models)

| Rank | Model | Configuration | Creator | Score |
|------|-------|---------------|---------|-------|
| 1 | [d1-3B](/models/d1-3b) | Liquid-run public split; official 0.2.1 scorer | LiquidAI | 56.4 skill points |
| 2 | [d1-omni-600M](/models/d1-omni-600m) | Liquid-run public split; official 0.2.1 scorer | LiquidAI | 12.9 skill points |

## FAQ

### What does Language (Liquid report) measure?

Liquid AI reports 56.4 skill points for d1-3B and 12.9 for d1-omni-600M on Language with the official Decision Index 0.2.1 scorer.

### Which model leads the published Language (Liquid report) snapshot?

d1-3B currently leads the published Language (Liquid report) snapshot with a score of 56.4 skill points.

### How many models are evaluated on Language (Liquid report)?

The October 7, 2026 provider report contains 2 AI models.

### Does Language (Liquid report) affect BenchLM's overall score?

Not directly. Language (Liquid report) is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.