FounderBench: leaderboard
FounderBench evaluates LLM agents in a deterministic simulator on repeated startup-like decisions using text-based structured actions. It spans 50 public tasks across 10 business task families. The pinned metric is Average Task Score, which measures the final task score mapped from simulator outcomes on a 0 to 100 scale. Source: BenchmarkList’s attributed result table.
Metric: Average Task Score (score). Source: benchmarklist.com. Saturation forecast: Around September 2027. 11 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.5 Flash | 67.69 | #66 |
| 2 | Grok 4.5 | 66.53 | #54 |
| 3 | GPT-5.6 Sol | 66.39 | #22 |
| 4 | Kimi K3 | 65.63 | #35 |
| 5 | Claude Sonnet 5 | 63.9 | #65 |
| 6 | Claude Sonnet 4.5 | 61.09 | #148 |
| 7 | DeepSeek V3 Chat | 56.59 | #457 |
| 8 | GLM 4.5 Air | 54.78 | #403 |
| 9 | Gemini 2.5 Flash | 52.69 | #255 |
| 10 | Grok 4.3 | 52.59 | #204 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=founderbench · How It Works · Data refreshed daily, snapshot 2026-11-05.