Compare Models
Select models (max 5)
Gemini 4 ArgonClaude Sonnet 5.5Claude Opus 5.5Claude Fable 5.1Claude Opus 5
 |  |  | | | |
|---|---|---|---|---|---|
| Vals IndexGDP-weighted benchmark | 68.90%±0.97 | 67.04%±0.92 | 66.97%±0.89 | 65.83%±1.11 | 63.67%±0.96 |
| Legal Research BenchAgentic US legal research | 54.81%±3.46 | 48.08%±3.47 | 50.48%±3.48 | 55.29%±3.46 | 55.29%±3.46 |
| Finance Agent (v2)Core financial analyst tasks | 65.40%±0.32 | 58.10%±0.67 | 58.59%±0.17 | 58.88%±2.06 | 58.63%±0.08 |
| Tax Agent BenchAgentic US corporate tax research | 76.23%±2.87 | 73.39%±2.98 | 70.50%±3.15 | 77.64%±2.83 | 75.06%±2.96 |
| MedCodeMedical billing code support | 58.80%±2.10 | 52.92%±2.12 | 49.80%±2.27 | 53.51%±2.17 | 63.57%±1.99 |
| Terminal-Bench ScienceExpert-authored scientific research workflows | 44.29%±5.98 | 38.57%±5.86 | 47.14%±6.01 | 40.00%±5.90 | 27.14%±5.35 |
| Code MigrationRewriting programs in new languages | 68.17%±4.35 | 69.83%±4.26 | 66.65%±4.33 | 54.61%±4.81 | 57.47%±4.37 |
| Terminal-Bench 4.0Frontier-difficulty terminal tasks | 57.58%±2.31 | 64.14%±1.01 | 65.15%±0.00 | 58.08%±3.31 | 53.53%±1.34 |
| Vibe Code Bench v1.1Building web apps from scratch | 91.91%±1.90 | 92.39%±1.26 | 90.29%±1.53 | 90.26%±1.57 | 88.40%±3.00 |
Benchmarks
Vals Index *
Gemini 4 Argon
0.00%± 0.97
(41/41)Claude Sonnet 5.5
0.00%± 0.92
(41/41)Claude Opus 5.5
0.00%± 0.89
(41/41)Claude Fable 5.1
0.00%± 1.11
(41/41)Claude Opus 5
0.00%± 0.96
(41/41)Legal Research Bench *
Gemini 4 Argon
0.00%± 3.46
(72/72)Claude Sonnet 5.5
0.00%± 3.47
(72/72)Claude Opus 5.5
0.00%± 3.48
(72/72)Claude Fable 5.1
0.00%± 3.46
(72/72)Claude Opus 5
0.00%± 3.46
(72/72)Finance Agent (v2) *
Gemini 4 Argon
0.00%± 0.32
(73/73)Claude Sonnet 5.5
0.00%± 0.67
(73/73)Claude Opus 5.5
0.00%± 0.17
(73/73)Claude Fable 5.1
0.00%± 2.06
(73/73)Claude Opus 5
0.00%± 0.08
(73/73)Tax Agent Bench *
Gemini 4 Argon
0.00%± 2.87
(64/64)Claude Sonnet 5.5
0.00%± 2.98
(64/64)Claude Opus 5.5
0.00%± 3.15
(64/64)Claude Fable 5.1
0.00%± 2.83
(64/64)Claude Opus 5
0.00%± 2.96
(64/64)MedCode *
Gemini 4 Argon
0.00%± 2.10
(104/104)Claude Sonnet 5.5
0.00%± 2.12
(104/104)Claude Opus 5.5
0.00%± 2.27
(104/104)Claude Fable 5.1
0.00%± 2.17
(104/104)Claude Opus 5
0.00%± 1.99
(104/104)Terminal-Bench Science
Gemini 4 Argon
0.00%± 5.98
(34/34)Claude Sonnet 5.5
0.00%± 5.86
(34/34)Claude Opus 5.5
0.00%± 6.01
(34/34)Claude Fable 5.1
0.00%± 5.90
(34/34)Claude Opus 5
0.00%± 5.35
(34/34)Code Migration *
Gemini 4 Argon
0.00%± 4.35
(71/71)Claude Sonnet 5.5
0.00%± 4.26
(71/71)Claude Opus 5.5
0.00%± 4.33
(71/71)Claude Fable 5.1
0.00%± 4.81
(71/71)Claude Opus 5
0.00%± 4.37
(71/71)Terminal-Bench 4.0
Gemini 4 Argon
0.00%± 2.31
(42/42)Claude Sonnet 5.5
0.00%± 1.01
(42/42)Claude Opus 5.5
0.00%± 0.00
(42/42)Claude Fable 5.1
0.00%± 3.31
(42/42)Claude Opus 5
0.00%± 1.34
(42/42)Vibe Code Bench v1.1 *
Gemini 4 Argon
0.00%± 1.90
(106/106)Claude Sonnet 5.5
0.00%± 1.26
(106/106)Claude Opus 5.5
0.00%± 1.53
(106/106)Claude Fable 5.1
0.00%± 1.57
(106/106)Claude Opus 5
0.00%± 3.00
(106/106)Overall performance
Performance on the Vals Index, a GDP-weighted aggregation of tasks across finance, coding, and law
•2026-09-30
Comparison by Industry
Model performance on different sections of the economy.
Legal
54.23%
vs.25.50%
vs.27.12%
vs.50.16%
vs.49.64%
Finance
72.29%
vs.69.07%
vs.68.34%
vs.71.99%
vs.70.89%
Healthcare
73.11%
vs.72.01%
vs.70.61%
vs.72.40%
vs.77.28%
Math
99.00%
vs.100.00%
vs.100.00%
vs.100.00%
vs.99.00%
Science
55.36%
vs.56.26%
vs.58.65%
vs.43.87%
vs.47.93%
Academic
N/A
vs.N/A
vs.N/A
vs.92.15%
vs.91.64%
Education
53.65%
vs.51.77%
vs.45.83%
vs.48.53%
vs.49.43%
Coding
64.03%
vs.77.35%
vs.61.00%
vs.60.10%
vs.62.42%
Cyber
61.07%
vs.44.87%
vs.44.47%
vs.46.66%
vs.38.79%
Social Mobility
69.76%
vs.67.19%
vs.70.64%
vs.74.90%
vs.76.93%
Cost Analysis
Per task and per token model pricing.
Vals Index · USD
$15.68
$21.34
$32.14
$28.71
$19.31
Cost / TestVals Index
$15.68
vs.$21.34
vs.$32.14
vs.$28.71
vs.$19.31
Input Cost/ 1M Tokens
$4.00
vs.$2.00
vs.$4.00
vs.$10.00
vs.$5.00
Input Cache Write/ 1M Tokens
N/A
vs.$2.50
vs.$5.00
vs.$12.50
vs.$6.25
Input Cache Read/ 1M Tokens
$0.20
vs.$0.20
vs.$0.40
vs.$0.25
vs.$0.50
Output Cost/ 1M Tokens
$20.00
vs.$10.00
vs.$20.00
vs.$50.00
vs.$25.00
Model Metadata
Basic information about each model.
| | | | | |
|---|---|---|---|---|---|
| Model provider | Google | Anthropic | Anthropic | Anthropic | Anthropic |
| Latency | 2792.77s | 4684.30s | 4740.12s | 4608.95s | 3461.38s |
| Cost (In/Out) | $4 / $20 | $2 / $10 | $4 / $20 | $10 / $50 | $5 / $25 |
| Context Window | 1M | 1M | 1M | 1M | 1M |
| Max Output Token | 262k | 128k | 128k | 128k | 128k |
| Input Modality |