Skip to content

Benchmark

FinanceAgent v1.1

Agentic benchmarks measure how well models carry out tasks that take several steps. Scores are percentages. Higher is better.

Models measured
11
Scores recorded
13
Sources
2
IQ weight
1.88

Score distribution

60%
Mean
60.1%
Median
4.2%
Std dev
54%
10th percentile
64.4%
90th percentile
53.7%
Min
53.7%69.2%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.88 for chat models, measured over 11 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Ling 3.0 Flash Fin· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim69.2%
2Claude Opus 4.7Vals AI external evaluation · adaptive thinking · high effort· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.64.4%
3Claude Sonnet 4.6Vals AI · max thinking· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.63.3%
4GPT-5.4 ProOpenAI GPT-5.4 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.61.5%
5Claude Sonnet 4.6Vals AI · high thinking· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.61.4%
6Doubao Seed 2.1 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim60.7%
7Nemotron 3 UltraNeMo Evaluator SDK · Vals.ai Financial Agent 1.1 · without web search· Reported by NVIDIA· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.60.1%
8GPT 5.5xhigh· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.60%
9GPT 5.2OpenAI GPT-5.4 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.59.5%
10Doubao Seed 2.1 Turbo· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim56%
11GPT 5.4OpenAI GPT-5.4 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.56%
12GPT-5.3 CodexOpenAI GPT-5.4 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.54%
13Nemotron 3 UltraNeMo Evaluator SDK · Vals.ai Financial Agent 1.1 · with web search· Reported by NVIDIA· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.53.7%

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Data by LLM Stats