Skip to main content

Benchmark profile

Vals Agent Poker Bench (Agent Poker Bench)

Which model can make the most money playing poker?

Data verified

How BenchLM shows Agent Poker Bench

BenchLM mirrors the public Vals AI Agent Poker Bench leaderboard captured from https://www.vals.ai/benchmarks/poker_agent and updated by Vals on December 23, 2025. The snapshot preserves overall scores, uncertainty, latency, cost-per-test metadata, and task-level scores where Vals publishes them.

Agent Poker Bench 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.

17 Vals rows1 task viewsprivate datasetTasks: OverallDisplay only

Agent Poker Bench score on Agent Poker Bench — December 23, 2025

BenchLM mirrors the published agent poker bench score view for Agent Poker Bench. GPT-5.2 leads the public snapshot at 1131.83% , followed by GPT-5 (1103.18%) and Gemini 3 Flash Preview (1100.21%). BenchLM does not use these results to rank models overall.

17 modelsExternal benchmark mirrorsCurrentDisplay onlyUpdated December 23, 2025

Agent Poker Bench score table (17 models)

Score
1
GPT-5.2OpenAI
1131.83%
2
GPT-5OpenAI
1103.18%
3
1100.21%
4
1090.30%
6
1078.90%
7
DeepSeek V3p1Fireworks
1058.23%
8
1055.50%
9
GPT-5.1OpenAI
1038.59%
11
1033.38%
12
1032.60%
13
GPT Oss 120bFireworks
1015.33%
14
Kimi K2 ThinkingMoonshot AI
1011.63%
15
994.51%
16
945.76%

The published Agent Poker Bench snapshot places GPT-5.2 first at 1131.83%. The third row is 31.62 points behind. The broader top-10 range is 97.53 points, so the table still separates the published systems.

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

About Agent Poker Bench

Year

2026

Tasks

Poker-playing agent trials

Format

Accuracy score

Difficulty

Strategic game-agent decision making

BenchLM mirrors the public Vals AI Agent Poker Bench 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.

BenchLM freshness & provenance

Version

Agent Poker Bench 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.

FAQ

What does Agent Poker Bench measure?

Which model can make the most money playing poker?

Which model leads the published Agent Poker Bench snapshot?

GPT-5.2 currently leads the published Agent Poker Bench snapshot with 1131.83% agent poker bench score. BenchLM shows this benchmark for display only and does not use it in overall rankings.

How many models are evaluated on Agent Poker Bench?

17 AI models are included in BenchLM's mirrored Agent Poker Bench snapshot, based on the public leaderboard captured on December 23, 2025.

Last updated: December 23, 2025 · mirrored from the public benchmark leaderboard

Choose a model with this week’s evidence

Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.

One email each week. Unsubscribe anytime.