Benchmarks · composite
Epoch Capabilities Index
Epoch AI's Capabilities Index: a single score per model fitted across many benchmarks (item-response model), so scores are comparable across models tested on different sets. About 100 is GPT-4.1 and about 150 is a current frontier model. Data: Epoch AI, CC BY.
180 models · 180 results · data as of 2026-10-29 · epoch.ai ↗
Leaderboard
Opus 5.5 max
All results
180 rows| # | Model | Effort | Epoch Capabilities Index |
|---|---|---|---|
| 1 | Opus 5.5 | max | 168.8 ±4.3 |
| 2 | GPT-6 Astra | max | 168.8 ±4.5 |
| 3 | GPT-6.1 Sol | max | 168.4 ±4.7 |
| 4 | Fable 5.1 | max | 166.5 ±3.9 |
| 5 | Gemini 4 Argon | max | 166.4 ±4.1 |
| 6 | Sonnet 5.5 | max | 166.2 ±4.1 |
| 7 | Opus 5 | max | 163.9 ±3.6 |
| 8 | GPT-6 Sol | max | 163.7 ±3.3 |
| 9 | Fable 5 | max | 163.4 ±3.8 |
| 10 | GPT-5.5 Pro | xhigh | 162.9 ±3.3 |
| 11 | GPT-5.6 Sol | max | 162.8 ±3.5 |
| 12 | Muse Wonder (Max) | max | 161.8 ±3.0 |
| 13 | GPT-5.6 Terra | max | 160.1 ±3.0 |
| 14 | GPT-5.5 | xhigh | 159.7 ±3.2 |
| 15 | GPT-5.4 Pro | xhigh | 159.6 ±2.8 |
| 16 | Haiku 5.5 | max | 159.3 ±3.1 |
| 17 | Opus 4.8 | max | 159.0 ±2.5 |
| 18 | Gemini 3.7 Flash | max | 158.0 ±2.7 |
| 19 | Kimi K3 | max | 157.9 ±2.8 |
| 20 | Gemini 3.8 Flash | max | 157.4 ±3.0 |
| 21 | Qwen 3.8 Max | max | 157.4 ±2.5 |
| 22 | GPT-5.4 | xhigh | 157.3 ±2.2 |
| 23 | Muse Spark 1.3 | max | 157.2 ±2.3 |
| 24 | GPT-5.3 Codex | xhigh | 157.1 ±3.9 |
| 25 | Sonnet 5 | max | 156.9 ±2.8 |
| 26 | Opus 4.7 | max | 156.8 ±2.1 |
| 27 | GPT-6 Luna | max | 156.8 ±2.6 |
| 28 | Grok 4.6 | max | 156.6 ±2.3 |
| 29 | GPT-5.6 Luna | max | 156.5 ±2.5 |
| 30 | GPT-5.2 Pro | xhigh | 156.3 ±3.0 |