# ComplexConstraints (Surge) (ComplexConstraints)

> An instruction-following benchmark from Surge AI that tests whether a model satisfies many simultaneous, interacting constraints in one response.

Canonical page: https://benchlm.ai/benchmarks/complexconstraints

- Category: [Instruction Following](/instruction-following)
- Last updated: November 13, 2026

## About ComplexConstraints

- Year: 2026
- Tasks: Prompts with multiple interacting constraints
- Format: Pass rate (%)
- Difficulty: Constrained instruction following
- Paper: [Introducing Inkling 2 Small](https://thinkingmachines.ai/news/inkling-2-small/)

Thinking Machines lists ComplexConstraints among the evaluations it takes from Surge and reports v1.0 scores in the Inkling 2 Small launch (November 5, 2026): DeepSeek-V4.1-Flash-1017 50.4%, Gemini 3.8 Flash 48.4%, Inkling 2 Small 44.5%, Muse Spark Open 41.3% and Kimi K3 37.9%.

ComplexConstraints is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (1 models)

The one result on this page was reported by the lab that built the model.

| Rank | Model | Creator | Score | Who ran it |
|------|-------|---------|-------|-----------|
| 1 | [Inkling 2 Small](/models/inkling-2-small) | Thinking Machines Lab | 44.5% | Reported by Thinking Machines Lab |

## FAQ

### What does ComplexConstraints measure?

An instruction-following benchmark from Surge AI that tests whether a model satisfies many simultaneous, interacting constraints in one response.

### Which model scores highest on ComplexConstraints?

Inkling 2 Small by Thinking Machines Lab currently leads with a score of 44.5% on ComplexConstraints.

### How many models are evaluated on ComplexConstraints?

1 AI models have published results on ComplexConstraints in the BenchLM catalog.

### Does ComplexConstraints affect BenchLM's overall score?

Not directly. ComplexConstraints is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.