Benchmark Almanac

Gemma 2 2B IT

Gemma 2 2B IT places in the top three on ALRAGE.

Google DeepMind Released July 31st, 2024 Family Gemma 2 The lab's model card Compare with others
23
Benchmarks with a score
0
Ranked first, out of five or more models
1
In the top three
Google DeepMind Jul 2024

Gemma 2 2B IT

Open weights Small

No Epoch Capabilities Index score yet
  • KOR-Bench 24.3% #30 /33
  • Encyclo-K 12.9% #23 /23
  • Alyah 26% #8 /10

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
ALRAGE 80.6% #2 of 39 Official leaderboard
Ko-IFEval (Upstage) 43.7% #4 of 21 Official leaderboard
MuSR 39.3% #63 of 98 Official leaderboard
Vectara Hallucination Leaderboard (HHEM) 7.00% #112 of 148 Official leaderboard
EQ-Bench v2 60.9 #66 of 85 Official leaderboard
WildBench 32.7 #37 of 47 Official leaderboard
Alyah 26% #8 of 10 Official leaderboard
Global PIQA 58.4% #36 of 44 Official leaderboard
LMArena (Elo) 1,200 #284 of 335 Official leaderboard
Open Arabic LLM Leaderboard v2 34.7% #35 of 39 Official leaderboard
AlGhafa 34.1% #36 of 39 Official leaderboard
ArabicMMLU 30.1% #36 of 39 Official leaderboard
MMLU-Pro 15.6% #201 of 212 Official leaderboard
ArbMMLU-HT 23.4% #37 of 39 Official leaderboard
SuperGPQA 18.6% #56 of 59 Official leaderboard
MadinahQA 20.1% #39 of 39 Official leaderboard
Encyclo-K 12.9% #23 of 23 Official leaderboard

Math and science

BenchmarkScoreRankSource
KOR-Bench 24.3% #30 of 33 Official leaderboard
ZebraLogic 4.20% #45 of 46 Official leaderboard

Coding

BenchmarkScoreRankSource
MHPP 9.50% #22 of 23 Official leaderboard

Safety and security

BenchmarkScoreRankSource
AraTrust 31.2% #36 of 39 Official leaderboard
nocot-bench (NCRI) 24.1 #184 of 185 Official leaderboard

Other

BenchmarkScoreRankSource
Open LLM Leaderboard v2 17.1 #65 of 99 Official leaderboard