Compare Models

Select models (max 7)

MiMo V2.5 ProDrex Max


Legal Research BenchAgentic US legal research	15.87%±2.54	N/A
Tax Agent BenchAgentic US corporate tax research	11.50%±1.96	N/A
MedCodeMedical billing code support	32.48%±1.91	N/A
Terminal-Bench 4.0Frontier-difficulty terminal tasks	0.51%±0.51	N/A
Vibe Code Bench v1.1Building web apps from scratch	34.11%±4.53	N/A

Benchmarks

Legal Research Bench *

MiMo V2.5 Pro

0.00%± 2.54

(84/84)

Drex Max

N/A

Tax Agent Bench *

MiMo V2.5 Pro

0.00%± 1.96

(75/75)

Drex Max

N/A

MedCode *

MiMo V2.5 Pro

0.00%± 1.91

(107/107)

Drex Max

N/A

Terminal-Bench 4.0

MiMo V2.5 Pro

0.00%± 0.51

(54/54)

Drex Max

N/A

Vibe Code Bench v1.1 *

MiMo V2.5 Pro

0.00%± 4.53

(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

MiMo V2.5 Pro Drex Max

Comparison by Industry

Model performance on different sections of the economy.

Legal

46.50%

vs.

N/A

Finance

42.65%

vs.

99.18%

Healthcare

58.11%

vs.

N/A

Math

22.00%

vs.

N/A

Academic

83.59%

vs.

N/A

Coding

47.49%

vs.

N/A

Cost Analysis

Per task and per token model pricing.

Vals Index · USD

N/A

Cost / TestVals Index

N/A

vs.

N/A

Input Cost/ 1M Tokens

$0.435

vs.

$0.06

Input Cache Read/ 1M Tokens

$0.004

vs.

N/A

Output Cost/ 1M Tokens

$0.87

vs.

$0.00

Model Metadata

Basic information about each model.


Model provider	Xiaomi	Decision
Latency	N/A	N/A
Cost (In/Out)	$0.435 / $0.87	$0.06 / $0
Context Window	1M	262k
Max Output Token	128k	0
Input Modality