ArchBench Serverless Component Generation: leaderboard
Architectural Component Generation: Generate complete serverless functions (FaaS) from specifications and codebase context.
Metric: Codebase Test Pass Rate (%, no human intervention). Source: github.com. Saturation forecast: Estimated already saturated. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4 | 79 | #438 |
| 2 | CodeQwen1.5-7B Chat | 69 | #1144 |
| 3 | GPT-3.5 Turbo | 68 | #865 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=archbench-serverless-component-generation · How It Works · Data refreshed daily, snapshot 2026-10-20.