Best Moonshot AI model overall
- Kimi K3 leads
- 43.6
- models ranked
- 7
- lead over #2
- +16.6
Overall index · read October 21
As of October 21, 2026, MoonshotAI Kimi K3 leads 7 models that qualify, with 43.6 on the overall index, ahead of MoonshotAI Kimi K2.6 (27.0) and MoonshotAI Kimi K2.7 Code (25.8).
| Rank | Model | Overall index | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|
| 1 | Kimi K3 Moonshot AI | 43.6 | 61.8 | $11.936 | |
| 2 | Kimi K2.6 Moonshot AI | 27.0 | 44.2 | $3.41 | |
| 3 | Kimi K2.7 Code Moonshot AI | 25.8 | 39.1 | $3.35 | |
| 4 | Kimi K2.5 Moonshot AI | 23.5 | 58.5 | $2.50 | |
| 5 | Kimi K2 Thinking Moonshot AI | 22.0 | 63.4 | $2.50 | |
| 6 | Kimi K2 0905 Moonshot AI | 15.3 | Not benchmarked | $2.50 | |
| 7 | Kimi K2 0711 Moonshot AI | 12.7 | 64.7 | $2.30 |
How this list is built
Generated from the dataset on every refresh. Nobody picks the order by hand.
Every Moonshot AI model on the overall index, ordered by the Artificial Analysis Intelligence Index, highest first. The index is a composite of evaluations Artificial Analysis runs itself, published on a 0 to 100 scale.
- OpenRouter, fetched
- Artificial Analysis (via OpenRouter), fetched