Compare Models
Select models (max 7)
Claude Sonnet 5Ember-1
 |  |  |
|---|---|---|
| Legal Research BenchAgentic US legal research | 41.83%±3.43 | 44.71%±3.46 |
| Tax Agent BenchAgentic US corporate tax research | 29.37%±3.25 | 31.08%±3.34 |
| MedCodeMedical billing code support | 47.54%±2.27 | N/A |
| Terminal-Bench-ScienceExpert-authored scientific research workflows | 5.71%±2.79 | 5.71%±2.79 |
| Terminal-Bench 4.0Frontier-difficulty terminal tasks | 10.61%±1.51 | 20.71%±3.64 |
| Vibe Code Bench v1.1Building web apps from scratch | 81.33%±3.05 | 83.87%±3.15 |
Benchmarks
Legal Research Bench *
Claude Sonnet 5
0.00%± 3.43
(84/84)Ember-1
0.00%± 3.46
(84/84)Tax Agent Bench *
Claude Sonnet 5
0.00%± 3.25
(75/75)Ember-1
0.00%± 3.34
(75/75)MedCode *
Claude Sonnet 5
0.00%± 2.27
(107/107)Ember-1
N/ATerminal-Bench-Science
Claude Sonnet 5
0.00%± 2.79
(42/42)Ember-1
0.00%± 2.79
(42/42)Terminal-Bench 4.0
Claude Sonnet 5
0.00%± 1.51
(54/54)Ember-1
0.00%± 3.64
(54/54)Vibe Code Bench v1.1 *
Claude Sonnet 5
0.00%± 3.05
(118/118)Ember-1
0.00%± 3.15
(118/118)Overall performance
Performance on the Vals Index, a GDP-weighted aggregation of tasks across finance, coding, and law
•2026-10-22
Comparison by Industry
Model performance on different sections of the economy.
Legal
62.87%
vs.44.71%
Finance
58.35%
vs.31.08%
Healthcare
61.80%
vs.N/A
Math
77.00%
vs.61.00%
Science
5.71%
vs.31.76%
Academic
86.48%
vs.N/A
Education
48.92%
vs.N/A
Coding
44.91%
vs.50.30%
Cyber
31.14%
vs.69.52%
Social Mobility
66.03%
vs.66.17%
Cost Analysis
Per task and per token model pricing.
Vals Index · USD
N/A
N/A
Cost / TestVals Index
N/A
vs.N/A
Input Cost/ 1M Tokens
$2.00
vs.$3.00
Input Cache Write/ 1M Tokens
$2.50
vs.N/A
Input Cache Read/ 1M Tokens
$0.20
vs.$0.30
Output Cost/ 1M Tokens
$10.00
vs.$15.00
Model Metadata
Basic information about each model.
| | |
|---|---|---|
| Model provider | Anthropic | Fireworks |
| Latency | N/A | N/A |
| Cost (In/Out) | $2 / $10 | $3 / $15 |
| Context Window | 1M | 1M |
| Max Output Token | 128k | 131k |
| Input Modality |