Samantha-Qwen-2-7B: benchmark results

macadeliccc's Samantha-persona conversational fine-tune of Qwen2 7B, QLoRA-trained on Opus Samantha and OpenHermes data. Provider: Other. Released 2024-06-15. Access: Open.

Unified ELO 1512 ± 1, rank #702 of 1553 rated models, from 7 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard - MuSR20.16Score95.1
Open Korean LLM Leaderboard41.99Average Score (%)81.2
Open LLM Leaderboard - MATH Level 521.15Score72.9
Open LLM Leaderboard - MMLU-Pro30.88Score64.2
Open LLM Leaderboard - BBH31.41Score56.9
Open LLM Leaderboard - IFEval43.77Score46.7
Open LLM Leaderboard - GPQA3.02Score27.6

Interactive version: theaggregate.ai/model?slug=samantha-qwen-2-7b · How It Works · Data refreshed daily, snapshot 2026-09-08.