Benchmark Almanac

Aya Expanse 32B

Aya Expanse 32B places in the top three on Alyah.

Cohere Released October 23rd, 2024 Family Aya Expanse The lab's model card Compare with others
22
Benchmarks with a score
0
Ranked first, out of five or more models
1
In the top three
Cohere Oct 2024

Aya Expanse 32B

Open weights Mid-size

No Epoch Capabilities Index score yet
  • KoBALT 25% #6 /12
  • Alyah 73.7% #3 /10
  • AlGhafa 77.6% #5 /39

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
AlGhafa 77.6% #5 of 39 Official leaderboard
ALRAGE 79.6% #7 of 39 Official leaderboard
Open Arabic LLM Leaderboard v2 67.2% #7 of 39 Official leaderboard
MadinahQA 53.5% #9 of 39 Official leaderboard
MLogiQA 53.8% #4 of 17 Official leaderboard
ArabicMMLU 60.6% #10 of 39 Official leaderboard
ArbMMLU-HT 58.9% #10 of 39 Official leaderboard
Alyah 73.7% #3 of 10 Official leaderboard
INCLUDE 59.5% #1 of 3 Official leaderboard
Global PIQA 71% #21 of 44 Official leaderboard
KoBALT 25% #6 of 12 Official leaderboard
BOUQuET 26.1 #11 of 18 Official leaderboard
Vectara Hallucination Leaderboard (2025 dataset) 10.9% #70 of 104 Official leaderboard
MMLU-Pro 45.4% #144 of 212 Official leaderboard
MuSR 38.7% #70 of 98 Official leaderboard
SEAL Multilingual 983 #14 of 19 Official leaderboard
LMArena (Elo) 1,267 #254 of 335 Official leaderboard
Vectara Hallucination Leaderboard (HHEM) 8.50% #124 of 148 Official leaderboard
Global MMLU Lite 73.5% #38 of 43 Official leaderboard

Math and science

BenchmarkScoreRankSource
DTBench 48.8% #168 of 180 Official leaderboard

Safety and security

BenchmarkScoreRankSource
AraTrust 89% #4 of 39 Official leaderboard

Other

BenchmarkScoreRankSource
Open LLM Leaderboard v2 29.7 #27 of 99 Official leaderboard