# GLM-5.1 vs Kimi K2.6

Every finance benchmark where both have a score (Zhipu AI and Moonshot AI). Data refreshed 2 Nov 2026. Scores from different benchmarks use different scales and are not added up.

| Benchmark | Metric | GLM-5.1 | GLM-5.1 source | Kimi K2.6 | Kimi K2.6 source | Ahead |
| --- | --- | --- | --- | --- | --- | --- |
| IPO Finance Agent | mean rubric score % (higher is better) | 79.7 | Paper Table 1, 30 Jun 2026, https://arxiv.org/html/2606.23032v3 | 68 | Paper Table 1, 30 Jun 2026, https://arxiv.org/html/2606.23032v3 | GLM-5.1 |
| KTD-Fin | total return % (higher is better) | -3.5 | Paper Table 4, 27 May 2026, https://arxiv.org/html/2605.28359v1 | -24.2 | Paper Table 4, 27 May 2026, https://arxiv.org/html/2605.28359v1 | GLM-5.1 |
| PortBench | CEPS (0-1) (higher is better) | 0.384 | Paper Table 2, 9 Oct 2026, https://arxiv.org/html/2605.27887v6 | 0.274 | Paper Table 2, 9 Oct 2026, https://arxiv.org/html/2605.27887v6 | GLM-5.1 |
| PortBench | mean QA score (0-1) (higher is better) | 0.757 | Paper Table 1, 9 Oct 2026, https://arxiv.org/html/2605.27887v6 | 0.511 | Paper Table 1, 9 Oct 2026, https://arxiv.org/html/2605.27887v6 | GLM-5.1 |

Source: https://benchmarks.finance/compare/glm-5-1-vs-kimi-k2-6/

Created and maintained by Moneda.