# Toolathlon → APEX-Agents-AA

Calibration · https://benchgap.net/calibration/1514 · data retrieved 2026-10-29

APEX-Agents-AA is estimated from Toolathlon with a Hill curve fitted on 15 models measured on both: y = 0.1361 + (1.2000 − 0.1361)·x^6.00 / (0.68010^6.00 + x^6.00), R² = 0.61, cross-validated error 7.6 pp. It is used for 0 estimates.

## Fitted on

| Model | Toolathlon | APEX-Agents-AA |
|---|---|---|
| [DeepSeek V4 Pro 0813](https://benchgap.net/model/deepseek-v4-pro-0813.md) | 51.8% | 24.3% |
| [Gemini 3.5 Flash](https://benchgap.net/model/gemini-3-5-flash.md) | 56.5% | 47.1% |
| [GLM-5](https://benchgap.net/model/glm-5.md) | 38.0% | 14.5% |
| [GLM-5.2](https://benchgap.net/model/glm-5-2.md) | 48.2% | 33.7% |
| [GPT-5.4](https://benchgap.net/model/gpt-5-4.md) | 54.6% | 33.3% |
| [GPT-5.4 mini](https://benchgap.net/model/gpt-5-4-mini.md) | 42.9% | 28.2% |
| [GPT-5.4 nano](https://benchgap.net/model/gpt-5-4-nano.md) | 35.5% | 24.9% |
| [GPT-5.5](https://benchgap.net/model/gpt-5-5.md) | 55.6% | 37.7% |
| [GPT-5.6 Luna](https://benchgap.net/model/gpt-5-6-luna.md) | 53.4% | 35.8% |
| [GPT-5.6 Terra](https://benchgap.net/model/gpt-5-6-terra.md) | 53.1% | 38.9% |
| [Kimi K2.6](https://benchgap.net/model/kimi-2-6.md) | 50.0% | 28.5% |
| [Kimi K2.5](https://benchgap.net/model/kimi-k2-5.md) | 27.8% | 11.5% |
| [MiniMax M2.7](https://benchgap.net/model/minimax-m2-7.md) | 46.3% | 10.6% |
| [Qwen3.5 397B](https://benchgap.net/model/qwen3-5-397b.md) | 36.3% | 15.3% |
| [Step 3.7 Flash](https://benchgap.net/model/step-3-7-flash.md) | 49.5% | 14.8% |