BhashaBench - Krishi - Hindi: leaderboard
From BharatGen's BhashaBench V1 leaderboard, which scores models on English and Hindi questions sourced from Indian government and domain-specific exams. BhashaBench-Krishi tests Indian agriculture knowledge; this board is accuracy on the Hindi-language questions only (%), higher is better. Flagged runs are skipped.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Around May 2027. 31 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o | 65.18 | #357 |
| 2 | Qwen 3 235B A22B 2507 Instruct | 64.13 | #315 |
| 3 | GPT-OSS-120B | 60.25 | #346 |
| 4 | Gemma 2 27B | 50.38 | #803 |
| 5 | DeepSeek V3.1 | 45.01 | #275 |
| 6 | Gemma 2 9B | 43.89 | #924 |
| 7 | Gemma 2 27B (IT) | 41.24 | #652 |
| 8 | Gemma 2 9B (IT) | 36.45 | #795 |
| 9 | GPT-OSS-20B | 36.27 | #523 |
| 10 | Llama 3.1 8B Instruct | 35.07 | #1017 |
| 11 | Pangea-7B | 34.71 | #1223 |
| 12 | Indic-gemma-7B-finetuned-sft-Navarasa-2.0 | 33.44 | #1372 |
| 13 | aya-23-8B | 33.22 | #1267 |
| 14 | Qwen 2.5 3B | 32.72 | #1177 |
| 15 | Llama 3.1 8B | 31.41 | #1130 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=bhashabench-krishi-hindi · How It Works · Data refreshed daily, snapshot 2026-11-05.