Hebrew LLM - Trivia (0-shot) — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 30 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 86.71 |
| 2 | Gemini 3 Flash (Preview) | 86.38 |
| 3 | Gemini 2.5 Pro | 82.06 |
| 4 | GPT-5.1 | 81.73 |
| 5 | GPT-4o | 77.08 |
| 6 | GLM-5.1 | 76.74 |
| 7 | Gemini 2.5 Flash | 76.41 |
| 8 | GPT-OSS-120B | 74.82 |
| 9 | Qwen 3.5 397B A17B | 72.09 |
| 10 | Qwen 3.5 122B A10B | 70.1 |
| 11 | GPT-5 Mini | 69.44 |
| 12 | Claude Sonnet 4.5 | 69 |
| 13 | NVIDIA-Nemotron-3-Super-120B-A12B-BF16 | 67.11 |
| 14 | jamba-large-1.7 | 66.11 |
| 15 | DeepSeek V3.1 | 64.45 |
Interactive version: aibenchmarks.dev/benchmark?slug=hebrew-llm-trivia-0-shot · How the rankings work · Data refreshed daily, snapshot 2026-07-12.