Infinity-Instruct-3M-0625-Qwen2-7B: benchmark results
BAAI's SFT-only fine-tune of Qwen2 7B on the Infinity-Instruct 3M and 0625 instruction datasets, tuned without RLHF (July 2024). Provider: BAAI. Released 2024-07-09. Access: Open.
Aggregate Rating 1467 ± 18, rank #901 of 1656 rated models, from 22 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GSM8K (flexible extract) | 62.93 | Exact match, flexible extract (%) | 95.7 |
| Open Korean LLM - Ko-EQ-Bench | 46.68 | EQ-Bench score | 94.3 |
| Open Ko-LLM Leaderboard | 44.9 | Average (%) | 91 |
| Open Korean LLM Leaderboard | 44.56 | Average Score (%) | 90.4 |
| Open Korean LLM - KorNAT-Helpful | 50.72 | Normalized accuracy (%) | 86.8 |
| Open Portuguese LLM - FaQuAD NLI | 79.19 | Macro F1 (%) | 86.7 |
| Open Korean LLM - Ko-IFEval | 41.42 | Strict accuracy, prompt/instruction mean (%) | 84.5 |
| Open Portuguese LLM - ENEM | 71.8 | Accuracy (%) | 81.5 |
| Open Portuguese LLM - ASSIN2 RTE | 92.6 | Macro F1 (%) | 80.9 |
| Open Portuguese LLM - BLUEX | 60.64 | Accuracy (%) | 78.7 |
| Open Portuguese LLM - OAB Exams | 50.43 | Accuracy (%) | 77 |
| Open LLM Leaderboard - MMLU-Pro | 39.6 | Score | 72.9 |
Interactive version: theaggregate.ai/model?slug=infinity-instruct-3m-0625-qwen2-7b · How It Works · Data refreshed daily, snapshot 2026-10-20.