Hebrew LLM - Summarization (Pairwise) — leaderboard
Metric: Pairwise Score (%). Source: huggingface.co. 30 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.1 | 65.98 |
| 2 | GPT-5 Mini | 55.11 |
| 3 | Claude Sonnet 4.5 | 55.07 |
| 4 | NVIDIA-Nemotron-3-Super-120B-A12B-BF16 | 52.81 |
| 5 | Gemini 3 Pro (Preview) | 51.76 |
| 6 | Gemini 3 Flash (Preview) | 49.75 |
| 7 | Gemini 2.5 Pro | 49.1 |
| 8 | MiniMax-M2.5 | 48.8 |
| 9 | Gemma 4 31B (IT) | 48.5 |
| 10 | GLM-5.1 | 47.9 |
| 11 | Qwen 3.5 397B A17B | 47.34 |
| 12 | Gemini 2.5 Flash | 46.94 |
| 13 | GLM-4.6 | 46.19 |
| 14 | Qwen 3.5 122B A10B | 46.19 |
| 15 | Gemma 4 26B A4B (IT) | 46.04 |
Interactive version: aibenchmarks.dev/benchmark?slug=hebrew-llm-summarization-pairwise · How the rankings work · Data refreshed daily, snapshot 2026-07-12.