Infinity-Instruct-7M-Gen-mistral-7B: benchmark results
BAAI's SFT-only fine-tune of Mistral 7B v0.1 on the Infinity-Instruct 7M foundational set plus its Gen chat stage, without RLHF (August 2024). Provider: BAAI. Released 2024-08-02. Access: Open.
Aggregate Rating 1455 ± 39, rank #3549 of 10916 rated models, from 16 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-IFEval | 44.85 | Strict accuracy, prompt/instruction mean (%) | 89.1 |
| AlpacaEval 2.0 | 39.67 | LC Win Rate (%) | 80.6 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 42.61 | Exact match, flexible extract (%) | 77.6 |
| Open Korean LLM - Ko-EQ-Bench | 28.45 | EQ-Bench score | 76.1 |
| Open Korean LLM Leaderboard | 40.27 | Average Score (%) | 76 |
| Open LLM Leaderboard - IFEval | 61.47 | Score | 73.6 |
| Open Korean LLM - KorNAT-Helpful | 48.43 | Normalized accuracy (%) | 67.9 |
| Open Korean LLM - KorNAT-Harmless | 62.99 | Normalized accuracy (%) | 53.2 |
| Open LLM Leaderboard - MuSR | 40.62 | Score | 47 |
| Open LLM Leaderboard - BBH | 49.64 | Score | 46.9 |
| Open LLM Leaderboard - MMLU-Pro | 32.74 | Score | 45.2 |
| Open LLM Leaderboard - GPQA | 29.03 | Score | 45 |
Interactive version: theaggregate.ai/model?slug=infinity-instruct-7m-gen-mistral-7b · How It Works · Data refreshed daily, snapshot 2026-11-08.