IndicBankBench - Wrong Info: leaderboard
The wrong-information axis of IndicBankBench (Paul et al., 2026): 142 retail-banking cases in which the customer states a plausible but incorrect claim that the assistant must check against the records. Percentage of cases passed in all three runs (strict pass^3), under deterministic safety and tool-use checks and a GLM-5.2 response judge; higher is better.
Metric: Strict pass^3 (%; 142 cases passed out of the 142 cases of the wrong-information axis (plausible but incorrect customer claims); each case run three times at temperature 0.7 with the stated reasoning modes; a case passes only when all three runs pass the deterministic safety and tool-use gates and the GLM-5.2 response judge; scripted Indian retail-banking conversations over deterministic mock tools). Source: arxiv.org. Saturation forecast: Around 2029. 11 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5 (Max) | 40.14 | #20 (Claude Opus 5) |
| 2 | Gemini 3.8 Flash | 38.03 | #36 |
| 3 | GLM-5.3 Flash (High) | 38.03 | #119 (GLM-5.3 Flash) |
| 4 | DeepSeek V4 Flash (0731) (Max) | 34.51 | #136 (DeepSeek V4 Flash (0731)) |
| 5 | DeepSeek V4 Pro (0813) (Max) | 32.39 | #81 (DeepSeek V4 Pro (0813)) |
| 6 | Grok 4.6 (High) | 30.28 | #44 (Grok 4.6) |
| 7 | MiniMax-M3 | 28.87 | #265 |
| 8 | Gemma 4 31B (IT) (Thinking) | 28.17 | #343 (Gemma 4 31B (IT)) |
| 9 | gemma-4-26B-A4B-it (Thinking) | 25.35 | |
| 10 | Gemma 4 E4B (IT) (Thinking) | 19.01 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=indicbankbench-wrong-info · How It Works · Data refreshed daily, snapshot 2026-11-12.