Compare Models
Select models (max 7)
Grok 4.6Kimi K3
| Vals IndexGDP-weighted benchmark | N/A | 52.95%±1.00 |
| Legal Research BenchAgentic US legal research | 48.08%±3.47 | 46.15%±3.46 |
| Finance Agent v2.1Core financial analyst tasks | N/A | 60.01%±2.09 |
| Tax Agent BenchAgentic US corporate tax research | 33.66%±3.41 | 33.37%±3.39 |
| MedCodeMedical billing code support | 44.71%±2.26 | 49.36%±2.20 |
| Terminal-Bench-ScienceExpert-authored scientific research workflows | 11.43%±3.83 | 1.43%±1.43 |
| Code Migration v1.1Rewriting programs in new languages | N/A | 28.59%±3.78 |
| Terminal-Bench 4.0Frontier-difficulty terminal tasks | 17.17%±1.34 | 17.17%±0.51 |
| Vibe Code Bench v1.1Building web apps from scratch | 76.24%±3.82 | 84.97%±2.75 |
Benchmarks
Vals Index *
Grok 4.6
N/AKimi K3
0.00%± 1.00
(41/41)Legal Research Bench *
Grok 4.6
0.00%± 3.47
(98/98)Kimi K3
0.00%± 3.46
(98/98)Finance Agent v2.1 *
Grok 4.6
N/AKimi K3
0.00%± 2.09
(44/44)Tax Agent Bench *
Grok 4.6
0.00%± 3.41
(89/89)Kimi K3
0.00%± 3.39
(89/89)MedCode *
Grok 4.6
0.00%± 2.26
(107/107)Kimi K3
0.00%± 2.20
(107/107)Terminal-Bench-Science
Grok 4.6
0.00%± 3.83
(47/47)Kimi K3
0.00%± 1.43
(47/47)Code Migration v1.1 *
Grok 4.6
N/AKimi K3
0.00%± 3.78
(30/30)Terminal-Bench 4.0
Grok 4.6
0.00%± 1.34
(68/68)Kimi K3
0.00%± 0.51
(68/68)Vibe Code Bench v1.1 *
Grok 4.6
0.00%± 3.82
(132/132)Kimi K3
0.00%± 2.75
(132/132)Overall performance
Performance on the Vals Index, a GDP-weighted aggregation of tasks across finance, coding, legal, and tax