ProLLM - Function Calling: leaderboard
ProLLM benchmark evaluating LLM function calling accuracy across practical API invocation tasks.
Metric: Score (%). Source: www.prollm.ai. Saturation forecast: Estimated already saturated. 90 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O1 | 97.3 | #183 |
| 2 | Gemini 2.0 Flash | 96.7 | #342 |
| 3 | GPT-4o Mini | 92.3 | #599 |
| 4 | Grok 3 Mini | 92.3 | #281 |
| 5 | GPT-4.5 | 91.9 | #274 |
| 6 | Grok Beta | 91.8 | #487 |
| 7 | Grok 3 | 91.1 | #304 |
| 8 | O3 Mini (Medium) | 91.1 | #272 (O3 Mini) |
| 9 | O3 Mini (High) | 91 | #272 (O3 Mini) |
| 10 | DeepSeek R1 | 90.4 | #251 |
| 11 | O1 Preview | 90 | #193 |
| 12 | GPT-4.1 | 89.9 | #248 |
| 13 | GPT-4.1 Mini | 89.8 | #355 |
| 14 | DeepSeek V3 | 89.8 | #321 |
| 15 | Mistral Small 3 | 89.6 | #627 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=prollm-function-calling · How It Works · Data refreshed daily, snapshot 2026-10-20.