BALSAM - Entailment: leaderboard
Metric: Overall score (0-100, LLM-judged generation and multiple choice). Source: benchmarks.ksaa.gov.sa. 29 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemma 4 31B (IT) | 100 |
| 2 | GPT-5.2 | 87.75 |
| 3 | Llama 4 Maverick Instruct | 86.92 |
| 4 | GLM-5 (Thinking) | 83 |
| 5 | Yehia-7B-preview | 79.67 |
| 6 | AceGPT-v2-8B-Chat | 78.83 |
| 7 | GLM-4.7 (Reasoning) | 78.42 |
| 8 | Llama 3.3 70B Instruct | 77.75 |
| 9 | Qwen 3 235B A22B | 77.08 |
| 10 | MiniMax M3 (Reasoning) | 76.92 |
| 11 | ALLaM-7B-Instruct-preview | 75.75 |
| 12 | Claude Opus 4.5 | 75.58 |
| 13 | GLM-5.2 | 75.17 |
| 14 | Jais-2-70B-Chat | 73.17 |
| 15 | Fanar-C-2-27B | 71.17 |
Interactive version: theaggregate.ai/benchmark?slug=balsam-entailment · How It Works · Data refreshed daily, snapshot 2026-09-19.