Compare Models
Select models (max 5)
GPT-5.6 LunaMiMo V2.6 Pro
 |  |  |
|---|---|---|
| Vals IndexGDP-weighted benchmark | 51.69%±1.06 | 55.20%±1.18 |
| Legal Research BenchAgentic US legal research | 36.54%±3.35 | 47.12%±3.47 |
| Finance Agent (v2)Core financial analyst tasks | 55.04%±0.31 | 57.34%±0.57 |
| Tax Agent BenchAgentic US corporate tax research | 60.81%±3.26 | 64.94%±3.21 |
| MedCodeMedical billing code support | 42.39%±2.27 | 44.97%±2.10 |
| Terminal-Bench ScienceExpert-authored scientific research workflows | 0.00%±0.00 | 2.86%±2.01 |
| Code MigrationRewriting programs in new languages | 44.55%±4.24 | 43.01%±4.32 |
| Terminal-Bench 4.0Frontier-difficulty terminal tasks | 11.62%±1.01 | 31.31%±3.07 |
| Vibe Code Bench v1.1Building web apps from scratch | 77.06%±3.07 | 85.22%±3.39 |
Benchmarks
Vals Index *
GPT-5.6 Luna
0.00%± 1.06
(43/43)MiMo V2.6 Pro
0.00%± 1.18
(43/43)Legal Research Bench *
GPT-5.6 Luna
0.00%± 3.35
(73/73)MiMo V2.6 Pro
0.00%± 3.47
(73/73)Finance Agent (v2) *
GPT-5.6 Luna
0.00%± 0.31
(74/74)MiMo V2.6 Pro
0.00%± 0.57
(74/74)Tax Agent Bench *
GPT-5.6 Luna
0.00%± 3.26
(65/65)MiMo V2.6 Pro
0.00%± 3.21
(65/65)MedCode *
GPT-5.6 Luna
0.00%± 2.27
(104/104)MiMo V2.6 Pro
0.00%± 2.10
(104/104)Terminal-Bench Science
GPT-5.6 Luna
0.00%± 0.00
(38/38)MiMo V2.6 Pro
0.00%± 2.01
(38/38)Code Migration *
GPT-5.6 Luna
0.00%± 4.24
(73/73)MiMo V2.6 Pro
0.00%± 4.32
(73/73)Terminal-Bench 4.0
GPT-5.6 Luna
0.00%± 1.01
(43/43)MiMo V2.6 Pro
0.00%± 3.07
(43/43)Vibe Code Bench v1.1 *
GPT-5.6 Luna
0.00%± 3.07
(108/108)MiMo V2.6 Pro
0.00%± 3.39
(108/108)Overall performance
Performance on the Vals Index, a GDP-weighted aggregation of tasks across finance, coding, and law
•2026-10-02
Comparison by Industry
Model performance on different sections of the economy.
Legal
40.61%
vs.28.97%
Finance
65.29%
vs.61.71%
Healthcare
63.39%
vs.66.64%
Math
60.00%
vs.70.00%
Science
25.30%
vs.9.09%
Academic
87.58%
vs.N/A
Education
44.22%
vs.45.05%
Coding
46.38%
vs.39.88%
Cyber
73.63%
vs.37.95%
Social Mobility
61.16%
vs.68.94%
Cost Analysis
Per task and per token model pricing.
Vals Index · USD
$0.824
$0.41
Cost / TestVals Index
$0.824
vs.$0.41
Input Cost/ 1M Tokens
$0.20
vs.$0.435
Input Cache Write/ 1M Tokens
$0.25
vs.N/A
Input Cache Read/ 1M Tokens
$0.02
vs.$0.004
Output Cost/ 1M Tokens
$1.20
vs.$0.87
Model Metadata
Basic information about each model.
| | |
|---|---|---|
| Model provider | OpenAI | Xiaomi |
| Latency | 1767.44s | 4038.37s |
| Cost (In/Out) | $0.2 / $1.2 | $0.435 / $0.87 |
| Context Window | 1M | 1M |
| Max Output Token | 128k | 128k |
| Input Modality |