Inference
EN
Vals AI

Public Benefits Bench v1.1

Can AI help people navigate SNAP benefits?

ModelsModels with a result.50
Top resultBest result on this benchmark.76.9%Claude Opus 5
Top-3 spreadPoints from first to third.3.2 pts
UpdatedDate the source changed the results.12 Oct 2026

Result and cost

50 of 50 results

Results

50 results
#ModelResultThe score from the source.CostUS dollars to run one task.TimeTime to run one task.
01Claude Opus 576.9% ±1.1$3.9530 min
02Claude Fable 5.174.9% ±1.1$7.4848 min
03Muse Wonder Full73.7% ±1.1$4.6473 min
04Claude Opus 5.570.6% ±1.2$5.2980 min
05Claude Fable 570.4% ±1.2$4.5921 min
06Muse Wonder70.2% ±1.2$2.4620 min
07Gemini 4 Argon69.8%$2.7926 min
08MiMo-V2.6-Pro68.9% ±1.2$0.08237 min
09Hy4 preview68.6% ±1.2$0.25363 min
10GLM-5.368.5% ±1.2$0.94736 min
10Muse Spark 1.268.5% ±1.2$0.5889 min
12Kimi K368.2% ±1.2$0.9137 min
13Claude Opus 4.868.1% ±1.2$1.8913 min
14MiMo-V2.6-Flash67.6% ±1.2$0.03221 min
15Claude Sonnet 5.567.2% ±1.2$3.7675 min
16Qwen3.8 Max67.1% ±1.2$0.99141 min
17Grok 4.666.8% ±1.2$0.91626 min
18GPT-5.6 Sol66.5% ±1.2$8.8561 min
19Ember-166.2% ±1.2$1.0213 min
20Claude Sonnet 566.0% ±1.2$1.2921 min
21Grok 4.765.6% ±1.2$2.2517 min
22Gemini 3.8 Flash65.3% ±1.2$1.013 min
23DeepSeek V4.1 Flash 101764.3% ±1.2$0.07310 min
24MiniMax M364.1% ±1.2$0.25412 min
25Claude Haiku 5.563.2% ±1.3$0.52537 min
26DeepSeek V4 Pro 042362.9% ±1.3$0.4615 min
27Claude Sonnet 4.662.4% ±1.3$1.1115 min
27GPT-5.6 Terra62.4% ±1.3$1.2020 min
29GLM-5.161.8% ±1.3$0.34110 min
30Muse Spark 1.161.3% ±1.3$0.2787 min
31GPT-5.6 Luna61.2% ±1.3$0.2739 min
32GPT-5.560.9% ±1.3$3.9740 min
33Gemini 3.5 Flash59.5% ±1.3$1.116 min
34Inkling-Small59.4% ±1.3$0.0969 min
35GPT-6.1 Sol59.3% ±1.3$0.69150 min
36Inkling58.5% ±1.3$0.36814 min
37Step 5 Preview58.1% ±1.3$0.50427 min
38GPT-6 Luna57.6% ±1.3$0.22453 min
39GPT-6 Sol56.6% ±1.3$2.4931 min
39Kimi K2.656.6% ±1.3$0.90610 min
39Gemini 3.6 Flash56.6% ±1.3$0.3852 min
42Claude Haiku 4.554.3% ±1.3$0.2253 min
43Gemini 3.1 Pro Preview53.8% ±1.3$0.3773 min
44Gemini 3.5 Flash-Lite51.8% ±1.3$0.09669 s
45Grok 4.351.7% ±1.3$0.2643 min
46Ling 3.0 Flash50.7% ±1.3$0.055 min
47Mercury 2.545.3% ±1.3$0.06868 s
48Grok 4.1 Fast (Reasoning)44.8% ±1.3$0.0312 min
49Laguna M.143.8% ±1.3$0.2026 min
50Laguna XS.241.4% ±1.3$0.1135 min