Compare Models
Select models (max 7)
DeepSeek V4.1 FlashKimi K3.1 Pro
 |  |  |
|---|---|---|
| Vals IndexGDP-weighted benchmark | N/A | 62.52%±1.14 |
| Legal Research BenchAgentic US legal research | 41.35%±3.42 | 37.50%±3.37 |
| Finance Agent v2.1Core financial analyst tasks | 65.43%±2.02 | 61.49%±2.10 |
| Tax Agent BenchAgentic US corporate tax research | 26.63%±3.14 | 24.87%±3.12 |
| MedCodeMedical billing code support | 41.17%±2.04 | N/A |
| Terminal-Bench-ScienceExpert-authored scientific research workflows | 0.00%±0.00 | N/A |
| Code Migration v1.1Rewriting programs in new languages | 58.94%±2.66 | N/A |
| Terminal-Bench 4.0Frontier-difficulty terminal tasks | 19.70%±1.75 | 63.64%±1.75 |
| Vibe Code Bench v1.1Building web apps from scratch | 84.74%±2.89 | 84.89%±2.95 |
Benchmarks
Vals Index *
DeepSeek V4.1 Flash
N/AKimi K3.1 Pro
0.00%± 1.14
(27/27)Legal Research Bench *
DeepSeek V4.1 Flash
0.00%± 3.42
(85/85)Kimi K3.1 Pro
0.00%± 3.37
(85/85)Finance Agent v2.1 *
DeepSeek V4.1 Flash
0.00%± 2.02
(31/31)Kimi K3.1 Pro
0.00%± 2.10
(31/31)Tax Agent Bench *
DeepSeek V4.1 Flash
0.00%± 3.14
(76/76)Kimi K3.1 Pro
0.00%± 3.12
(76/76)MedCode *
DeepSeek V4.1 Flash
0.00%± 2.04
(107/107)Kimi K3.1 Pro
N/ATerminal-Bench-Science
DeepSeek V4.1 Flash
0.00%± 0.00
(42/42)Kimi K3.1 Pro
N/ACode Migration v1.1 *
DeepSeek V4.1 Flash
0.00%± 2.66
(28/28)Kimi K3.1 Pro
N/ATerminal-Bench 4.0
DeepSeek V4.1 Flash
0.00%± 1.75
(55/55)Kimi K3.1 Pro
0.00%± 1.75
(55/55)Vibe Code Bench v1.1 *
DeepSeek V4.1 Flash
0.00%± 2.89
(119/119)Kimi K3.1 Pro
0.00%± 2.95
(119/119)Overall performance
Performance on the Vals Index, a GDP-weighted aggregation of tasks across finance, coding, and law
•2026-10-24
Comparison by Industry
Model performance on different sections of the economy.
Legal
62.31%
vs.24.58%
Finance
53.27%
vs.52.12%
Healthcare
63.34%
vs.N/A
Math
54.00%
vs.N/A
Science
29.65%
vs.N/A
Education
47.88%
vs.N/A
Coding
42.10%
vs.74.27%
Cyber
37.23%
vs.N/A
Social Mobility
64.28%
vs.N/A
Cost Analysis
Per task and per token model pricing.
Vals Index · USD
N/A
$6.44
Cost / TestVals Index
N/A
vs.$6.44
Input Cost/ 1M Tokens
$0.30
vs.$2.00
Input Cache Read/ 1M Tokens
$0.006
vs.$0.20
Output Cost/ 1M Tokens
$1.20
vs.$10.00
Model Metadata
Basic information about each model.
| | |
|---|---|---|
| Model provider | DeepSeek | Kimi |
| Latency | N/A | 3816.45s |
| Cost (In/Out) | $0.3 / $1.2 | $2 / $10 |
| Context Window | 1M | 1M |
| Max Output Token | 384k | 1M |
| Input Modality |