Sindhi LLM Benchmark: leaderboard
Sindhi-language question answering on 500 scored items, with the publisher averaging repeated runs. Reports correct answers as a percentage of scored questions; unparsed responses remain in the denominator. This board reports Sindhi LLM Benchmark using Accuracy (%).
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Around February 2027. 15 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5 | 95.4 | #16 |
| 2 | Gemini 3.6 Flash | 95.2 | #57 |
| 3 | Claude Opus 4.7 | 91.2 | #51 |
| 4 | DeepSeek V4.1 Flash | 89.29 | #37 |
| 5 | Claude Opus 4.8 | 89 | #39 |
| 6 | GPT-5.6 Luna | 88.4 | #79 |
| 7 | Claude Sonnet 4.6 | 85 | #94 |
| 8 | Claude Sonnet 5 | 84.4 | #65 |
| 9 | GLM-5.2 | 83.2 | #80 |
| 10 | Gemma 4 31B | 76.4 | #212 |
| 11 | GPT-OSS-120B | 69.2 | #346 |
| 12 | Qwen 3 32B | 63.8 | #449 |
| 13 | Claude Haiku 4.5 | 60.4 | #287 |
| 14 | GPT-OSS-20B | 56.45 | #523 |
| 15 | Nova Pro | 53.8 | #594 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=sindhi-llm-benchmark · How It Works · Data refreshed daily, snapshot 2026-11-05.