| Kimi K3 | Claude Fable 5.2 | |
|---|---|---|
| CTO Index | 55.6 (#27) | 95.5 (#1) |
| Provider | Moonshot AI | Anthropic |
| Weights | Open weights | Closed |
| Price $/M in / out | $3 / $15 | $8 / $40 |
| Context window | 1M | 1M |
| License | Kimi K3 License | Proprietary |
| Terminal-Bench 4.1 | 23.0% | 65.8% |
| WebDev Arena | 1659 | 1824 |
| ARC-AGI-2 | 60.4% | 93.8% |
| Text Arena | 1488 | 1503 |
| ARC-AGI-3 | — | 81.2% |
Snapshot 22 November 2026 · Scores copied from each benchmark's own leaderboard · How the CTO Index works
Frequently asked questions
Is Kimi K3 better than Claude Fable 5.2?
Claude Fable 5.2 scores higher on the CTO Index (95.5 vs 55.6). Across the 4 benchmarks both have results, Kimi K3 wins 0 and Claude Fable 5.2 wins 4 (Terminal-Bench, WebDev, ARC-AGI-2, Text Arena). Kimi K3 costs 63% less per blended token.
Which is better for coding agents, Kimi K3 or Claude Fable 5.2?
On Terminal-Bench 4.1, Claude Fable 5.2 scores 65.8% against Kimi K3's 23.0%. Differences under about three points are within the error bars.
How do Kimi K3 and Claude Fable 5.2 compare on price?
Kimi K3 lists at $3 / $15 and Claude Fable 5.2 at $8 / $40 per million input / output tokens.