Skip to main content
BenchLM

Vals Public Benefits Bench v1 (Public Benefits Bench v1)

We show this table for reference; we do not rank on it.

Data verified 36 confirmed releases in the last 30 daysFollow model changes

Can AI help people navigate SNAP benefits?

Public Benefits Bench v1 score on Public Benefits Bench v1 — June 9, 2026

We mirror the published public benefits bench v1 score view for Public Benefits Bench v1. Claude Fable 5 leads the public snapshot at 71.65%, followed by Claude Opus 4.8 (62.11%) and MiniMax M3 (60.69%). We do not use these results to rank models overall.

13 modelsAgenticCurrentDisplay onlyUpdated June 9, 2026

Public Benefits Bench v1 score table (13 models)

Score
1
Claude Fable 5Anthropic · Closed
71.65%
2
Claude Opus 4.8Anthropic · Closed
62.11%
3
MiniMax M3MiniMax · Open weight
60.69%
4
Claude Sonnet 4.6Anthropic · Closed
58.52%
5
Gemini 3.5 FlashGoogle · Closedhigh reasoning
57.98%
6
GLM-5.1Z.AI · Open weight
57.92%
7
DeepSeek V4 Pro 0813DeepSeek · Open weightmax reasoning
57.58%
8
GPT-5.5OpenAI · Closedxhigh reasoning
57.24%
9
Kimi K2.6Moonshot AI · Open weight
53.38%
10
Gemini 3.1 Pro PreviewGooglehigh reasoning
50.88%
11
Grok 4.3xAI · Closedhigh reasoning
50.07%
12
Claude Haiku 4.5Anthropic · Closed
49.53%
13
44.32%

How Public Benefits Bench v1 is shown here

BenchLM mirrors the public Vals AI Public Benefits Bench v1 leaderboard captured from https://www.vals.ai/benchmarks/public-benefits-bench-v1 and updated by Vals on June 9, 2026. The snapshot preserves overall scores, uncertainty, latency, cost-per-test metadata, and task-level scores where Vals publishes them.

Public Benefits Bench v1 is display only on BenchLM. Vals proprietary or Vals-hosted aggregate views are useful context, but BenchLM does not use them as weighted ranking inputs or as a replacement for benchmark-native source records.

Snapshot

13 Vals rows5 task viewsprivate datasetTasks: Overall, Both, Web Search, Multi Turn, NeitherDisplay only

The published Public Benefits Bench v1 snapshot places Claude Fable 5 first at 71.65%. The third row is 10.96 points behind. The broader top-10 range is 20.77 points, so the table still separates the published systems.

13 models have been evaluated on Public Benefits Bench v1. The benchmark falls in the Agentic category. We keep external benchmark mirrors separate from the weighted global scoring system, so these results remain source-specific evidence. Public Benefits Bench v1 is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About Public Benefits Bench v1

Year

2026

Tasks

SNAP public-benefits navigation tasks

Format

Accuracy score

Difficulty

Public-benefits policy navigation

BenchLM mirrors the public Vals AI Public Benefits Bench v1 leaderboard as display-only external evidence. The captured snapshot preserves overall scores, task-level scores where Vals publishes them, uncertainty, latency, and cost-per-test metadata. It is excluded from BenchLM weighted rankings.

Freshness and provenance

Version

Public Benefits Bench v1 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

Questions

What does Public Benefits Bench v1 measure?

Can AI help people navigate SNAP benefits?

Which model leads the published Public Benefits Bench v1 snapshot?

Claude Fable 5 currently leads the published Public Benefits Bench v1 snapshot with 71.65% public benefits bench v1 score. BenchLM shows this benchmark for display only and does not use it in overall rankings.

How many models are evaluated on Public Benefits Bench v1?

The June 9, 2026 snapshot contains 13 AI models.

Last updated: June 9, 2026 · mirrored from the public benchmark leaderboard

Know when it’s worth switching models

The model to choose, the cheaper alternative, and the release we would wait on.

Read a sample issue

Join 2,000+ readers.

One email each week. Unsubscribe anytime.