Calibration
AA Agentic Index → MCP Atlas
MCP Atlas is estimated from AA Agentic Index with a offset logistic curve fitted on 23 models measured on both: y = 0.6456 + (0.7870 − 0.6456) / (1 + exp(−153.26·(x − 0.2328))), R² = 0.49, cross-validated error 7.8 pp. It is used for 30 estimates.
| Estimated model | AA Agentic Index | MCP Atlas | Source |
|---|---|---|---|
| Qwen3.8 Max Preview | 49.6% | 78.7% | estimated ± 7.8 pp, low confidence |
| Muse Spark 1.2 | 44.0% | 78.7% | estimated ± 7.8 pp, low confidence |
| Quasar 438B | 32.7% | 78.7% | estimated ± 7.8 pp, low confidence |
| Gemini 3.6 Flash | 30.2% | 78.7% | estimated ± 7.8 pp, low confidence |
| Hy3 | 25.6% | 78.3% | estimated ± 7.8 pp, low confidence |
| Hy3 Preview | 25.6% | 78.3% | estimated ± 7.8 pp, low confidence |
| Nemotron 3 Ultra | 21.7% | 65.7% | estimated ± 7.8 pp, low confidence |
| Qwen3.6-27B | 20.1% | 64.7% | estimated ± 7.8 pp, low confidence |
| Grok 4.3 | 17.2% | 64.6% | estimated ± 7.8 pp, low confidence |
| MiniMax M2.7 | 16.8% | 64.6% | estimated ± 7.8 pp, low confidence |
| Gemini 3.5 Flash-Lite | 15.9% | 64.6% | estimated ± 7.8 pp, low confidence |
| Gemini 3.1 Pro | 10.3% | 64.6% | estimated ± 7.8 pp, low confidence |
| Qwen3.5-122B-A10B | 9.6% | 64.6% | estimated ± 7.8 pp, low confidence |
| Mistral Medium 3.5 128B | 9.4% | 64.6% | estimated ± 7.8 pp, low confidence |
| GPT-OSS 120B | 6.2% | 64.6% | estimated ± 7.8 pp, low confidence |
| Nemotron 3.5 Lightning 30B A3B NVFP4 | 6.1% | 64.6% | estimated ± 7.8 pp, low confidence |
| Granite 4.2 8B | 3.7% | 64.6% | estimated ± 7.8 pp, low confidence |
| Command A+ | 3.6% | 64.6% | estimated ± 7.8 pp, low confidence |
| Gemini 2.5 Pro | 3.6% | 64.6% | estimated ± 7.8 pp, low confidence |
| Mistral Large 3 | 2.4% | 64.6% | estimated ± 7.8 pp, low confidence |
| Mistral Small 4 | 1.4% | 64.6% | estimated ± 7.8 pp, low confidence |
| GPT-OSS 20B | 1.4% | 64.6% | estimated ± 7.8 pp, low confidence |
| Nemotron 3 Nano 30B | 1.0% | 64.6% | estimated ± 7.8 pp, low confidence |
| Trinity-Large-Preview | 1.2% | 64.6% | estimated ± 7.8 pp, low confidence |
| Trinity-Large-Thinking | 1.2% | 64.6% | estimated ± 7.8 pp, low confidence |
| Celeris-1 | 0.7% | 64.6% | estimated ± 7.8 pp, low confidence |
| DeepSeek V3 | 0.8% | 64.6% | estimated ± 7.8 pp, low confidence |
| Gemma 3 27B | 0.1% | 64.6% | estimated ± 7.8 pp, low confidence |
| Llama 4 Maverick | 0.6% | 64.6% | estimated ± 7.8 pp, low confidence |
| Llama 4 Scout | 0.5% | 64.6% | estimated ± 7.8 pp, low confidence |
Fitted on
| Model | AA Agentic Index | MCP Atlas |
|---|---|---|
| Claude Opus 4.7 | 39.5% | 77.3% |
| Claude Opus 4.8 | 42.6% | 82.2% |
| Claude Opus 5 | 56.2% | 85.8% |
| DeepSeek V4 Flash 0731 | 41.7% | 69.0% |
| DeepSeek V4 Pro | 49.6% | 73.6% |
| Gemini 3.5 Flash | 27.3% | 83.6% |
| GLM-5.1 | 25.2% | 71.8% |
| GLM-5.2 | 39.4% | 76.8% |
| GPT-5.4 mini | 19.7% | 57.7% |
| GPT-5.4 nano | 17.7% | 56.1% |
| GPT-5.5 | 37.3% | 75.3% |
| Inkling | 24.3% | 74.1% |
| Inkling-Small | 25.0% | 79.6% |
| Kimi K2.6 | 22.1% | 55.9% |
| Kimi K2.7 Code | 22.5% | 76.0% |
| Kimi K3 | 50.6% | 84.2% |
| Ling 3.0 Flash | 21.0% | 65.5% |
| MiniMax M3 | 30.8% | 74.2% |
| Muse Glimmer 30B | 10.5% | 75.5% |
| Muse Spark 1.1 | 27.5% | 88.1% |
| Qwen3.6-35B-A3B | 15.0% | 62.8% |
| Qwen3.7 Max | 23.9% | 76.4% |
| Qwen3.7 Plus | 19.7% | 73.2% |