Agentic benchmark
AA Tau3 Banking leaderboard
Artificial Analysis Tau3-Banking. Every model the catalog carries a published AA Tau3 Banking value for, ranked by that value.
CategoryAgentic
MeasureTask success rate
TasksBanking tool-use workflows
DifficultyAgentic banking workflows
An independently evaluated Tau3 banking benchmark from Artificial Analysis.
AA Tau3 Banking ranking
15 models with a published AA Tau3 Banking value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Task success rate |
|---|---|---|---|
| 1 |  Grok 4.6 | xAI | 50.7 |
| 2 |  Muse Spark 1.3 | Meta | 50.5 |
| 3 |  GLM-5.3 | Z.AI | 50.3 |
| 4 |  Qwen3.8-27B | Alibaba | 48.0 |
| 5 |  Claude Fable 5.1 | Anthropic | 47.2 |
| 5 | GLM-5.3-Flash | Z.AI | 47.2 |
| 7 |  Kimi K3 | Moonshot AI | 46.0 |
| 8 |  Gemini 3.8 Flash | 44.9 | |
| 9 |  GPT-5.6 Sol | OpenAI | 44.3 |
| 10 | Claude Opus 5 | Anthropic | 42.1 |
| 11 | GPT-6 Astra | OpenAI | 41.4 |
| 12 | GPT-5.6 Terra | OpenAI | 40.2 |
| 13 |  DeepSeek V4 Pro 0813 | DeepSeek | 39.6 |
| 14 | Claude Fable 5 | Anthropic | 38.1 |
| 15 |  Ling 3.0 Flash | InclusionAI | 28.0 |
Evidence key: Observed
Rows are ordered by the value Artificial Analysis Tau3-Banking Benchmark Leaderboard published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.