Nexus Function Calling - Overall: leaderboard
Nexus Function Calling mean exact-match accuracy across all nine primary API task sets, including Heldout Combined. Only models with all nine scores receive an overall; individual task results remain available for incomplete cohorts.
Metric: Mean Nine-Task Function Call Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | NexusRaven-V2-13B | 58.2 | #1382 |
| 2 | GPT-4 Preview (1106) | 54.18 | #418 |
| 3 | GPT-3.5 Turbo (1106) | 37.45 | #961 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=nexus-function-calling-overall · How It Works · Data refreshed daily, snapshot 2026-10-20.