Compare Models

Select models (max 5)

Grok 4.7Muse Spark 1.3


Vals IndexGDP-weighted benchmark	54.95%±1.07	53.20%±1.07
Legal Research BenchAgentic US legal research	47.12%±3.47	40.87%±3.42
Finance Agent (v2)Core financial analyst tasks	52.25%±0.35	58.90%±0.44
Tax Agent BenchAgentic US corporate tax research	65.60%±3.24	71.93%±2.92
MedCodeMedical billing code support	49.55%±2.17	N/A
Terminal-Bench ScienceExpert-authored scientific research workflows	10.00%±3.61	4.29%±2.44
Code MigrationRewriting programs in new languages	44.82%±4.21	27.58%±4.07
Terminal-Bench 4.0Frontier-difficulty terminal tasks	28.79%±2.31	10.61%±1.75
Vibe Code Bench v1.1Building web apps from scratch	86.17%±2.18	82.86%±2.90

Benchmarks

Vals Index *

Grok 4.7

0.00%± 1.07

(43/43)

Muse Spark 1.3

0.00%± 1.07

(43/43)

Legal Research Bench *

Grok 4.7

0.00%± 3.47

(73/73)

Muse Spark 1.3

0.00%± 3.42

(73/73)

Finance Agent (v2) *

Grok 4.7

0.00%± 0.35

(74/74)

Muse Spark 1.3

0.00%± 0.44

(74/74)

Tax Agent Bench *

Grok 4.7

0.00%± 3.24

(65/65)

Muse Spark 1.3

0.00%± 2.92

(65/65)

MedCode *

Grok 4.7

0.00%± 2.17

(104/104)

Muse Spark 1.3

N/A

Terminal-Bench Science

Grok 4.7

0.00%± 3.61

(38/38)

Muse Spark 1.3

0.00%± 2.44

(38/38)

Code Migration *

Grok 4.7

0.00%± 4.21

(73/73)

Muse Spark 1.3

0.00%± 4.07

(73/73)

Terminal-Bench 4.0

Grok 4.7

0.00%± 2.31

(43/43)

Muse Spark 1.3

0.00%± 1.75

(43/43)

Vibe Code Bench v1.1 *

Grok 4.7

0.00%± 2.18

(108/108)

Muse Spark 1.3

0.00%± 2.90

(108/108)

Overall performance

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

•2026-10-02

Grok 4.7 Muse Spark 1.3

Comparison by Industry

Model performance on different sections of the economy.

Legal

48.00%

vs.

31.89%

Finance

61.61%

vs.

64.52%

Healthcare

69.47%

vs.

N/A

Math

26.00%

vs.

55.00%

Science

34.83%

vs.

4.29%

Education

40.79%

vs.

N/A

Coding

43.60%

vs.

33.10%

Cyber

69.46%

vs.

69.41%

Social Mobility

65.63%

vs.

N/A

Cost Analysis

Per task and per token model pricing.

Vals Index · USD

$12.12

$3.376

Cost / TestVals Index

$12.12

vs.

$3.376

Input Cost/ 1M Tokens

$2.00

vs.

$1.25

Input Cache Read/ 1M Tokens

$0.50

vs.

$0.15

Output Cost/ 1M Tokens

$6.00

vs.

$4.25

Model Metadata

Basic information about each model.


Model provider	SpaceXAI	Meta
Latency	2131.40s	1962.38s
Cost (In/Out)	$2 / $6	$1.25 / $4.25
Context Window	500k	1M
Max Output Token	500k	131k
Input Modality