Open Portuguese LLM - ASSIN2 RTE: leaderboard
Recognising textual entailment between Brazilian Portuguese sentence pairs from ASSIN 2. From the final table of the archived Open Portuguese LLM Leaderboard (15-shot, lm-evaluation-harness), covering about 1,210 open checkpoints and 42 API models; macro F1 (%), higher is better. Frozen since the last evaluation on 2025-09-01.
Metric: Macro F1 (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 1253 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 2.5 72B Instruct | 95.1 | #469 |
| 2 | GPT-5 (Minimal) | 94.98 | #105 (GPT-5) |
| 3 | GPT-5 Nano | 94.93 | #439 |
| 4 | DeepSeek V3.1 | 94.93 | #275 |
| 5 | Qwen 2 72B Instruct | 94.93 | #579 |
| 6 | Qwen 2 57B A14B Instruct | 94.93 | #744 |
| 7 | EZO-Qwen2.5-32B-Instruct | 94.89 | #484 |
| 8 | GPT-5 Mini (2025-08-07) | 94.87 | #175 |
| 9 | Llama 3.1 405B Instruct | 94.76 | #470 |
| 10 | Qwen 2.5 32B Instruct | 94.73 | #509 |
| 11 | Qwen 2.5 VL 72B Instruct | 94.73 | #383 |
| 12 | Claude 3.7 Sonnet (20250219) | 94.73 | #209 |
| 13 | Qwen 2.5 VL 32B Instruct | 94.73 | #463 |
| 14 | Qwen2.5-14B-Instruct-1M | 94.73 | #581 |
| 15 | Qwen2.5-14B-Gutenberg-Instruct-Slerpeno | 94.73 | #431 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=open-portuguese-llm-assin2-rte · How It Works · Data refreshed daily, snapshot 2026-11-05.