# MCP Atlas → Toolathlon-Verified

Calibration · https://benchgap.net/calibration/mcpatlas/current/to/toolathlonverified/current · data retrieved 2026-11-18

Toolathlon-Verified is estimated from MCP Atlas with a linear curve fitted on 12 models measured on both: y = 0.9519·x + -0.0706, R² = 0.60, cross-validated error 8.8 pp. It is used for 2 estimates.

| Estimated model | MCP Atlas | Toolathlon-Verified | Source |
|---|---|---|---|
| [Ripple-65B Preview](https://benchgap.net/model/ripple-65b-preview.md) | 71.7% | 61.2% | estimated ± 8.8 pp, medium confidence |
| [LLaDA2.2-flash](https://benchgap.net/model/llada2-2-flash.md) | 46.2% | 36.9% | estimated ± 8.8 pp, low confidence |

## Fitted on

| Model | MCP Atlas | Toolathlon-Verified |
|---|---|---|
| [Claude Opus 5](https://benchgap.net/model/claude-opus-5.md) | 85.8% | 80.6% |
| [DeepSeek V4 Flash 0731](https://benchgap.net/model/deepseek-v4-flash-0731.md) | 69.0% | 70.3% |
| [DeepSeek V4 Pro](https://benchgap.net/model/deepseek-v4-pro.md) | 73.6% | 74.1% |
| [ERNIE 5.5 Flash Preview](https://benchgap.net/model/ernie-5-5-flash-preview.md) | 84.6% | 75.0% |
| [Hy4 preview](https://benchgap.net/model/hy4-preview.md) | 83.7% | 74.1% |
| [Inkling-Small](https://benchgap.net/model/inkling-small.md) | 79.6% | 54.4% |
| [Kimi K3](https://benchgap.net/model/kimi-k3.md) | 84.2% | 73.2% |
| [Muse Spark Open](https://benchgap.net/model/muse-spark-open.md) | 89.2% | 72.8% |
| [Ornith-1.5-35B-A3B](https://benchgap.net/model/ornith-1-5-35b-a3b.md) | 70.2% | 48.7% |
| [Ornith-1.5-397B](https://benchgap.net/model/ornith-1-5-397b.md) | 80.0% | 71.2% |
| [Ornith-1.5-9B](https://benchgap.net/model/ornith-1-5-9b.md) | 54.2% | 41.2% |
| [Step 5 Preview](https://benchgap.net/model/step-5-preview.md) | 85.6% | 74.1% |
