---
title: "Public Benefits Bench - Benchmark Evaluation - FinanceBenchmark"
description: "Evaluates civic policy compliance, public benefits eligibility, and regulatory adjudication."
canonical: "https://financebenchmark.ai/benchmarks/public-benefits-bench"
type: "text/markdown"
---
# Public Benefits Bench

> Evaluates civic policy compliance, public benefits eligibility, and regulatory adjudication.

## Table of Contents
- [1. Specifications & Metadata](#specifications)
- [2. Model Rankings on this Benchmark](#rankings)
- [3. Source Attribution & Citation](#attribution)

## 1. Specifications & Metadata {#specifications}

- **Slug:** `public-benefits-bench`
- **Domain Categories:** verification, forensic
- **Composite Weight:** 10 (Tier 3)
- **Task Count:** Variable
- **Source / Publisher:** VALS (2026)

## 2. Model Rankings on this Benchmark {#rankings}

| Rank | Model | Organization | Overall | Eval Date | Source |
|:----:|:------|:-------------|:---:|:---------:|:------:|
| 1 | [Claude Opus 5](https://financebenchmark.ai/models/claude-opus-5.md) | Anthropic | **76.9** | 2026-07 | [Ref](https://vals.ai) |
| 2 | [Claude Fable 5](https://financebenchmark.ai/models/claude-fable-5.md) | Anthropic | **70.4** | 2026-06 | [Ref](https://vals.ai) |
| 3 | [Muse Spark 1.2](https://financebenchmark.ai/models/muse-spark-1-2.md) | Meta | **68.5** | 2026-08 | [Ref](https://vals.ai) |
| 4 | [Kimi K3](https://financebenchmark.ai/models/kimi-k3.md) | Moonshot AI | **68.3** | 2026-07 | [Ref](https://vals.ai) |
| 5 | [Qwen 3.8 Max](https://financebenchmark.ai/models/qwen-3-8-max.md) | Alibaba | **67.1** | 2026-08 | [Ref](https://vals.ai) |
| 6 | [Grok 4.6](https://financebenchmark.ai/models/grok-4-6.md) | xAI | **66.8** | 2026-08 | [Ref](https://vals.ai) |
| 7 | [GPT-5.6 Sol](https://financebenchmark.ai/models/gpt-5-6-sol.md) | OpenAI | **66.5** | 2026-07 | [Ref](https://vals.ai) |
| 8 | [Claude Sonnet 5](https://financebenchmark.ai/models/claude-sonnet-5.md) | Anthropic | **66.0** | 2026-06 | [Ref](https://vals.ai) |
| 9 | [GPT-5.6 Terra](https://financebenchmark.ai/models/gpt-5-6-terra.md) | OpenAI | **62.4** | 2026-07 | [Ref](https://vals.ai) |
| 10 | [Muse Spark 1.1](https://financebenchmark.ai/models/muse-spark-1-1.md) | Meta | **61.3** | 2026-07 | [Ref](https://vals.ai) |
| 11 | [GPT-5.6 Luna](https://financebenchmark.ai/models/gpt-5-6-luna.md) | OpenAI | **61.2** | 2026-07 | [Ref](https://vals.ai) |
| 12 | [Inkling Small](https://financebenchmark.ai/models/inkling-small.md) | Thinking Machines | **59.4** | 2026-07 | [Ref](https://vals.ai) |
| 13 | [Inkling](https://financebenchmark.ai/models/inkling.md) | Thinking Machines | **58.5** | 2026-07 | [Ref](https://vals.ai) |
| 14 | [Gemini 3.6 Flash](https://financebenchmark.ai/models/gemini-3-6-flash.md) | Google | **56.6** | 2026-07 | [Ref](https://vals.ai) |
| 15 | [Gemini 3.5 Flash Lite](https://financebenchmark.ai/models/gemini-3-5-flash-lite.md) | Google | **51.8** | 2026-07 | [Ref](https://vals.ai) |
| 16 | [Ling 3.0 Flash](https://financebenchmark.ai/models/ling-3-0-flash.md) | Ant Group | **50.7** | 2026-07 | [Ref](https://vals.ai) |

## 3. Source Attribution & Citation {#attribution}

All scores for Public Benefits Bench are derived directly from published literature or reproducible evaluation harnesses.
To submit updated evaluations, email [submit@financebenchmark.ai](mailto:submit@financebenchmark.ai).