Benchmark Almanac

Qwen2-7B-Instruct

Qwen2-7B-Instruct places in the top three on Ko-IFEval (Upstage).

Qwen Released June 7th, 2024 Family Qwen2 The lab's model card Compare with others
20
Benchmarks with a score
0
Ranked first, out of five or more models
1
In the top three
Qwen Jun 2024

Qwen2-7B-Instruct

Open weights Small

No Epoch Capabilities Index score yet
  • KOR-Bench 30.7% #29 /33
  • TOMG-Bench 0.15% #5 /5
  • BigCodeBench-Hard 11.8% #80 /96

Where it ranks

Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.

Knowledge and language

BenchmarkScoreRankSource
Ko-IFEval (Upstage) 46.9% #2 of 21 Official leaderboard
MadinahQA 59.5% #6 of 39 Official leaderboard
AlGhafa 73.2% #8 of 39 Official leaderboard
Open Arabic LLM Leaderboard v2 63.6% #9 of 39 Official leaderboard
Open Arabic LLM Leaderboard v1 51.7% #13 of 55 Official leaderboard
ArabicMMLU 60% #11 of 39 Official leaderboard
ArbMMLU-HT 51.3% #13 of 39 Official leaderboard
CFBench 50.2% #8 of 19 Official leaderboard
ALRAGE 71.1% #21 of 39 Official leaderboard
MMLU-CF 58.3% #21 of 37 Official leaderboard
MuSR 39.3% #64 of 98 Official leaderboard
MMLU-Pro 47.2% #140 of 212 Official leaderboard

Math and science

BenchmarkScoreRankSource
MathBench 53.4% #8 of 21 Official leaderboard
KOR-Bench 30.7% #29 of 33 Official leaderboard
ZebraLogic 8.40% #41 of 46 Official leaderboard

Coding

BenchmarkScoreRankSource
BigCodeBench 35.6% #54 of 69 Official leaderboard
BigCodeBench-Hard 11.8% #80 of 96 Official leaderboard

Safety and security

BenchmarkScoreRankSource
AraTrust 82.8% #11 of 39 Official leaderboard
TOMG-Bench 0.15% #5 of 5 Official leaderboard

Other

BenchmarkScoreRankSource
Open LLM Leaderboard v2 27.9 #29 of 99 Official leaderboard