The AI arena is free today

Open Superagent

Tau3 Banking

Progress Over Time

Interactive timeline showing model performance evolution on Tau3 Banking

State-of-the-art frontier
Open
Proprietary

Tau3 Banking Leaderboard

22 models
ContextCostLicense
1124B262K$0.06 / $0.18
2
Thinking Machines Lab
Thinking Machines Lab
594B256K$0.30 / $1.00
3
Reflection AI
Reflection AI
501B1.0M$0.40 / $1.20
4
Institute of Foundation Models
Institute of Foundation Models
375B——
5—500K$2.00 / $6.00
6———
7
Inception
Inception
—1.0M$0.10 / $0.35
8
InclusionAI
InclusionAI
124B131K$0.06 / $0.18
9
Institute of Foundation Models
Institute of Foundation Models
36B——
10
Institute of Foundation Models
Institute of Foundation Models
7B——
11
Institute of Foundation Models
Institute of Foundation Models
32B——
12
Thinking Machines Lab
Thinking Machines Lab
975B524K$0.95 / $4.05
1330B131K$0.30 / $1.20
14—524K$0.30 / $1.20
15120B262K$0.20 / $1.00
16
Institute of Foundation Models
Institute of Foundation Models
4B——
1774B——
18
Thinking Machines Lab
Thinking Machines Lab
276B524K$0.30 / $1.20
19128B256K$1.50 / $7.50
204B——
2130B262K$0.08 / $0.20
22
Liquid AI
Liquid AI
3B——
Notice missing or incorrect data?
About this benchmark

What is Tau3 Banking?

τ³-Bench banking domain evaluates agentic models on multi-turn, tool-using customer-support scenarios in a simulated retail banking environment.

Tau3 Banking is a text benchmark evaluating models on reasoning, agents, and tool calling tasks. LLM Stats tracks 22 models on this benchmark, scored on a 0–1 scale. The current average is 0.2, with the leader at 0.4.

Compare leaders on the best AI for reasoning, best AI for agents and best AI for tool calling leaderboards.

Current leaders

Ling 3.0 Flash Fin from InclusionAI currently leads the Tau3 Banking leaderboard with a score of 0.410 across 22 evaluated AI models.

1Ling 3.0 Flash FinInclusionAI41.0%
2Inkling 2 SmallThinking Machines Lab40.4%
3Beam 1.0Reflection AI36.1%

FAQ

Common questions about the Tau3 Banking benchmark and leaderboard.

What is the Tau3 Banking benchmark?

τ³-Bench banking domain evaluates agentic models on multi-turn, tool-using customer-support scenarios in a simulated retail banking environment.

What is the Tau3 Banking leaderboard?

The Tau3 Banking leaderboard ranks 22 AI models based on their performance on this benchmark. Currently, Ling 3.0 Flash Fin by InclusionAI leads with a score of 0.410. The average score across all models is 0.242.

What is the highest Tau3 Banking score?

The highest Tau3 Banking score is 0.410, achieved by Ling 3.0 Flash Fin from InclusionAI.

How many models are evaluated on Tau3 Banking?

22 models have been evaluated on the Tau3 Banking benchmark, with 0 verified results and 21 self-reported results.

What categories does Tau3 Banking cover?

Tau3 Banking is categorized under reasoning, agents, and tool calling. The benchmark evaluates text models.

What is the best open-source model on Tau3 Banking?

Beam 1.0 by Reflection AI is the top-ranked open-source model on Tau3 Banking, with a score of 0.361 (rank #3).

Which model offers the best value on Tau3 Banking?

Among models scoring within 10% of the leader, Ling 3.0 Flash Fin from InclusionAI is the cheapest, at $0.06 per million input tokens with a score of 0.410.

How recent are the Tau3 Banking leaderboard results?

The Tau3 Banking leaderboard was last updated in November 2026 and currently includes 22 evaluated models.