Compare Models

Select models (max 7)

Qwen 3.8 Max 0902Drex Max


Vals IndexGDP-weighted benchmark	59.05%±1.08	N/A
Legal Research BenchAgentic US legal research	46.15%±3.46	N/A
Finance Agent v2.1Core financial analyst tasks	58.83%±2.12	N/A
Tax Agent BenchAgentic US corporate tax research	49.66%±3.60	N/A
Code Migration v1.1Rewriting programs in new languages	57.19%±2.80	N/A
Terminal-Bench 4.0Frontier-difficulty terminal tasks	38.89%±1.34	N/A
Vibe Code Bench v1.1Building web apps from scratch	77.14%±3.72	N/A

Benchmarks

Vals Index *

Qwen 3.8 Max 0902

0.00%± 1.08

(26/26)

Drex Max

N/A

Legal Research Bench *

Qwen 3.8 Max 0902

0.00%± 3.46

(84/84)

Drex Max

N/A

Finance Agent v2.1 *

Qwen 3.8 Max 0902

0.00%± 2.12

(30/30)

Drex Max

N/A

Tax Agent Bench *

Qwen 3.8 Max 0902

0.00%± 3.60

(75/75)

Drex Max

N/A

Code Migration v1.1 *

Qwen 3.8 Max 0902

0.00%± 2.80

(28/28)

Drex Max

N/A

Terminal-Bench 4.0

Qwen 3.8 Max 0902

0.00%± 1.34

(54/54)

Drex Max

N/A

Vibe Code Bench v1.1 *

Qwen 3.8 Max 0902

0.00%± 3.72

(118/118)

Drex Max

N/A

Overall performance

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

•2026-10-22

Qwen 3.8 Max 0902 Drex Max

Comparison by Industry

Model performance on different sections of the economy.

Legal

30.79%

vs.

N/A

Finance

60.16%

vs.

99.18%

Coding

49.51%

vs.

N/A

Cost Analysis

Per task and per token model pricing.

Vals Index · USD

$8.73

N/A

Cost / TestVals Index

$8.73

vs.

N/A

Input Cost/ 1M Tokens

$2.00

vs.

$0.06

Input Cache Write/ 1M Tokens

$2.50

vs.

N/A

Input Cache Read/ 1M Tokens

$0.25

vs.

N/A

Output Cost/ 1M Tokens

$6.00

vs.

$0.00

Model Metadata

Basic information about each model.


Model provider	Alibaba	Decision
Latency	6345.62s	N/A
Cost (In/Out)	$2 / $6	$0.06 / $0
Context Window	984k	262k
Max Output Token	128k	0
Input Modality