FullStackBench en: leaderboard
FullStackBench executable code generation across 16 programming languages and 11 main application domains. This source collection is labelled English; its historical scores also include bilingual runs, so legacy results are not uniformly English-only.
Metric: Score (imported). Source: arxiv.org. Source: benchmarklist.com. Saturation forecast: Around May 2027. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.5 122B A10B | 62.6 | #173 |
| 2 | Qwen 3.5 27B | 60.1 | #187 |
| 3 | DeepSeek V2.5 | 58.65 | #489 |
| 4 | Qwen 3.5 35B A3B | 58.1 | #256 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=fullstackbench-en · How It Works · Data refreshed daily, snapshot 2026-10-20.