Benchmark Almanac

Aya Expanse 8B

Aya Expanse 8B.

Cohere Released October 23rd, 2024 Family Aya Expanse The lab's model card Compare with others
21
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Cohere Oct 2024

Aya Expanse 8B

Open weights Small

No Epoch Capabilities Index score yet
  • KoBALT 19% #9 /12
  • Alyah 61.2% #6 /10
  • Open Arabic LLM Leaderboard v1 53.3% #10 /55

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
Open Arabic LLM Leaderboard v1 53.3% #10 of 55 Official leaderboard
Open Arabic LLM Leaderboard v2 60.9% #12 of 39 Official leaderboard
ArabicMMLU 57.5% #13 of 39 Official leaderboard
AlGhafa 66.7% #14 of 39 Official leaderboard
ArbMMLU-HT 49.2% #14 of 39 Official leaderboard
ALRAGE 75.8% #16 of 39 Official leaderboard
MadinahQA 48.7% #16 of 39 Official leaderboard
Vectara Hallucination Leaderboard (2025 dataset) 9.50% #53 of 104 Official leaderboard
Alyah 61.2% #6 of 10 Official leaderboard
Global PIQA 63.9% #27 of 44 Official leaderboard
INCLUDE 47.8% #2 of 3 Official leaderboard
KoBALT 19% #9 of 12 Official leaderboard
MLogiQA 36.9% #13 of 17 Official leaderboard
MuSR 37.3% #77 of 98 Official leaderboard
MMLU-Pro 33.7% #173 of 212 Official leaderboard
LMArena (Elo) 1,223 #275 of 335 Official leaderboard
BOUQuET 21.7 #15 of 18 Official leaderboard
Vectara Hallucination Leaderboard (HHEM) 12.2% #136 of 148 Official leaderboard

Math and science

BenchmarkScoreRankSource
DTBench 45.1% #177 of 180 Official leaderboard

Safety and security

BenchmarkScoreRankSource
AraTrust 82.5% #12 of 39 Official leaderboard

Other

BenchmarkScoreRankSource
Open LLM Leaderboard v2 22.4 #49 of 99 Official leaderboard