Capability
Indexes · 2026-08-03
Synthetic general capability score from the Atlas factor model. Higher scores indicate stronger estimated overall benchmark performance.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Mythos Preview | 166.3 |
| GPT-5.6 Sol Pro | 164.9 |
| Claude Mythos 5 | 164.2 |
| Claude Fable 5 | 164.2 |
| Claude Opus 5 | 163.8 |
| GPT-5.5 Pro | 162.9 |
| GPT-5.6 Terra Pro | 162.8 |
| GPT-5.6 Sol | 162.6 |
| Gemini 3 Deep Think | 161.4 |
| GPT-5.6 Luna Pro | 161.3 |
| GPT-5.4 Pro | 161.2 |
| GPT-5.6 Terra | 160.9 |
| Kimi K3 | 160.4 |
| GPT-5.5 | 160.3 |
| Opus 4.8 | 160.2 |
| GPT-5.6 Luna | 159.0 |
| Grok 4.5 | 158.9 |
| DeepSeek-V4-Flash-0731 | 158.6 |
| Sonnet 5 | 158.4 |
| Opus 4.7 | 158.4 |
| Muse Spark 1.1 | 158.0 |
| GPT-5.4 | 157.7 |
| Opus 4.6 | 156.8 |
| Gemini 3.5 Flash | 156.8 |
| GPT-5.3-Codex | 156.7 |
| GPT-5.2 Pro | 156.6 |
| GLM-5.2 | 156.5 |
| Gemini 3.6 Flash | 156.2 |
| Gemini 3.1 Pro Preview | 155.1 |
| Laguna S 2.1 | 155.1 |
| Gemini 2.5 Deep Think | 154.8 |
| Sonnet 4.6 | 154.6 |
| Muse Spark | 154.6 |
| Hy3 | 154.5 |
| Motif-3-Beta | 154.4 |
| Composer 2.5 | 154.4 |
| GPT-5.2 | 154.2 |
| Nex-N2-Pro | 154.2 |
| Qwen3.7-Max | 154.1 |
| GPT-5.2-Codex | 153.8 |