Core RankingsCompare
| Claude Fable 5by Anthropic | Claude Opus 4.8by Anthropic | |
|---|---|---|
| Evaluation | ||
| Core fitHigher is better | 64 | 63 |
| AA Intelligence IndexHigher is better | 60 | 56 |
| Class | Frontier | Frontier |
| Rated effort | max | max |
| Coding agents | ||
| Coding Agent IndexHigher is better | 66 | 61 |
| DeepSWEHigher is better | 66 | 56 |
| Terminal-Bench v2Higher is better | 83 | 79 |
| SWE-Atlas-QnAHigher is better | 49 | 47 |
| Coding $ / taskLower is better | $11.71 | $7.70 |
| Minutes / taskLower is better | 23.4 | 23.1 |
| Cost | ||
| AA $ / taskLower is better | $2.75 | $1.80 |
| Input $ / 1MLower is better | $10.00 | $5.00 |
| Output $ / 1MLower is better | $50.00 | $25.00 |
| Cache read $ / 1MLower is better | $1.00 | $0.50 |
Core fit is Core's dated product judgment; the AA index and task cost are published by Artificial Analysis; coding results name their harness. The better value in each measured row is emphasized — rows without two measurements name no winner.