Compare Models
Select models (max 7)
Claude Fable 5.2GPT-6.2 AstraGPT-6.2 Sol
| Vals IndexGDP-weighted benchmark | 71.31%±0.96 | 67.66%±0.94 | 66.06%±0.91 |
| Legal Research BenchAgentic US legal research | 46.15%±3.46 | 42.79%±3.44 | 37.02%±3.36 |
| Finance Agent v2.1Core financial analyst tasks | 69.47%±1.98 | 60.75%±2.12 | 58.91%±2.10 |
| Tax Agent BenchAgentic US corporate tax research | 32.09%±3.37 | 21.84%±2.79 | 24.84%±3.12 |
| Terminal-Bench-ScienceExpert-authored scientific research workflows | 55.71%±5.98 | 71.43%±5.44 | 58.57%±5.93 |
| Code Migration v1.1Rewriting programs in new languages | 83.27%±3.12 | N/A | N/A |
| Terminal-Bench 4.0Frontier-difficulty terminal tasks | 65.15%±2.31 | 63.13%±1.34 | 60.61%±0.00 |
| Vibe Code Bench v1.1Building web apps from scratch | 93.77%±1.15 | 89.90%±2.04 | 87.62%±2.19 |
Benchmarks
Vals Index *
Claude Fable 5.2
0.00%± 0.96
(41/41)GPT-6.2 Astra
0.00%± 0.94
(41/41)GPT-6.2 Sol
0.00%± 0.91
(41/41)Legal Research Bench *
Claude Fable 5.2
0.00%± 3.46
(98/98)GPT-6.2 Astra
0.00%± 3.44
(98/98)GPT-6.2 Sol
0.00%± 3.36
(98/98)Finance Agent v2.1 *
Claude Fable 5.2
0.00%± 1.98
(44/44)GPT-6.2 Astra
0.00%± 2.12
(44/44)GPT-6.2 Sol
0.00%± 2.10
(44/44)Tax Agent Bench *
Claude Fable 5.2
0.00%± 3.37
(89/89)GPT-6.2 Astra
0.00%± 2.79
(89/89)GPT-6.2 Sol
0.00%± 3.12
(89/89)Terminal-Bench-Science
Claude Fable 5.2
0.00%± 5.98
(47/47)GPT-6.2 Astra
0.00%± 5.44
(47/47)GPT-6.2 Sol
0.00%± 5.93
(47/47)Code Migration v1.1 *
Claude Fable 5.2
0.00%± 3.12
(30/30)GPT-6.2 Astra
N/AGPT-6.2 Sol
N/ATerminal-Bench 4.0
Claude Fable 5.2
0.00%± 2.31
(68/68)GPT-6.2 Astra
0.00%± 1.34
(68/68)GPT-6.2 Sol
0.00%± 0.00
(68/68)Vibe Code Bench v1.1 *
Claude Fable 5.2
0.00%± 1.15
(132/132)GPT-6.2 Astra
0.00%± 2.04
(132/132)GPT-6.2 Sol
0.00%± 2.19
(132/132)Overall performance
Performance on the Vals Index, a GDP-weighted aggregation of tasks across finance, coding, legal, and tax
Comparison by Industry
Model performance on different sections of the economy.
Legal
41.71%
vs.29.52%
vs.24.55%
Finance
59.65%
vs.52.87%
vs.52.68%
Math
100.00%
vs.N/A
vs.N/A
Science
63.74%
vs.68.44%
vs.57.76%
Coding
64.70%
vs.69.39%
vs.74.11%
Cost Analysis
Per task and per token model pricing.
Vals Index · USD
$17.94
$10.68
$3.179
Cost / TestVals Index
$17.94
vs.$10.68
vs.$3.179
Input Cost/ 1M Tokens
$8.00
vs.$8.00
vs.$2.00
Input Cache Write/ 1M Tokens
$10.00
vs.$10.00
vs.$2.50
Input Cache Read/ 1M Tokens
$0.20
vs.$0.40
vs.$0.10
Output Cost/ 1M Tokens
$40.00
vs.$40.00
vs.$10.00
Model Metadata
Basic information about each model.
| Model provider | |||
| Latency | 2621.82s | 3136.31s | 2506.87s |
| Cost (In/Out) | $8 / $40 | $8 / $40 | $2 / $10 |
| Context Window | 1M | 1M | 1M |
| Max Output Token | 128k | 128k | 128k |
| Input Modality |