Beyond Function Calling: leaderboard
Large language models are increasingly deployed as agents that solve tasks by interacting with external tool environments. Although recent tool-use benchmarks increasingly cover complex task settings, they still largely assume clean, stable, and trustworthy tool environments, leaving tool-environment unreliability insufficiently examined. We introduce ToolBench-X, a benchmark for evaluating agents
Metric: Overall (self-reported). Source: benchmarklist.com. Saturation forecast: Around September 2027. 12 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Seed 2.0 Lite | 51.3 | #145 |
| 2 | GPT-5.4 | 45.3 | #79 |
| 3 | DeepSeek V4 Pro | 42.5 | #102 |
| 4 | GLM-5.1 | 42 | #97 |
| 5 | Qwen 3.5 35B A3B (Thinking) | 41.9 | #256 (Qwen 3.5 35B A3B) |
| 6 | Gemini 3.1 Flash Lite | 41.6 | #191 |
| 7 | Claude Sonnet 4.6 | 41 | #90 |
| 8 | Qwen 3.5 35B A3B | 37.2 | #256 |
| 9 | GPT-4o | 35.9 | #343 |
| 10 | MiniMax-M2.7 | 24.9 | #258 |
| 11 | Kimi K2.6 | 21.2 | #104 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=beyond-function-calling · How It Works · Data refreshed daily, snapshot 2026-10-20.