Compare Models

Select models (max 7)
DeepSeek V4 Flash 0731Gemini 3.5 Flash
Benchmarks

Legal Research Bench *

DeepSeek V4 Flash 0731
0.00%± 3.19
(98/98)
Gemini 3.5 Flash
0.00%± 3.21
(98/98)

Tax Agent Bench *

DeepSeek V4 Flash 0731
0.00%± 3.16
(89/89)
Gemini 3.5 Flash
0.00%± 2.96
(89/89)

MedCode *

DeepSeek V4 Flash 0731
0.00%± 2.15
(107/107)
Gemini 3.5 Flash
0.00%± 2.11
(107/107)

Terminal-Bench-Science

DeepSeek V4 Flash 0731
0.00%± 2.44
(47/47)
Gemini 3.5 Flash
0.00%± 2.79
(47/47)

Terminal-Bench 4.0

DeepSeek V4 Flash 0731
0.00%± 3.54
(68/68)
Gemini 3.5 Flash
0.00%± 1.51
(68/68)

Vibe Code Bench v1.1 *

DeepSeek V4 Flash 0731
0.00%± 3.44
(132/132)
Gemini 3.5 Flash
0.00%± 4.73
(132/132)

Overall performance

Performance on the Vals Index, a GDP-weighted aggregation of tasks across finance, coding, legal, and tax

2026-11-16

Comparison by Industry

Model performance on different sections of the economy.

Legal
54.00%
57.19%
Finance
49.38%
54.71%
Healthcare
60.89%
66.20%
Math
56.00%
31.00%
Science
34.37%
5.71%
Academic
88.05%
90.15%
Coding
50.41%
45.65%
Cyber
72.08%
N/A
Social Mobility
N/A
61.22%

Cost Analysis

Per task and per token model pricing.

Vals Index · USD
N/A
N/A
Cost / TestVals Index
N/A
N/A
Input Cost/ 1M Tokens
$0.30
$1.50
Input Cache Read/ 1M Tokens
$0.006
$0.15
Output Cost/ 1M Tokens
$1.20
$9.00

Model Metadata

Basic information about each model.

Model provider
DeepSeekDeepSeek
GoogleGoogle
LatencyN/AN/A
Cost (In/Out)$0.3 / $1.2$1.5 / $9
Context Window1M1M
Max Output Token384k66k
Input Modality