Ahma-2-4B-Instruct: benchmark results

Provider: Other. Access: Open.

Aggregate Rating 1375 ± 41, rank #7027 of 10916 rated models, from 15 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
EuroEval Finnish NLU - Scandisent FI91.13Matthews correlation coefficient (%)74.2
EuroEval Estonian NLU - Estonian Valence45.69Matthews correlation coefficient (%)52.4
EuroEval Estonian NLU - Grammar ET6.59Matthews correlation coefficient (%)46.2
EuroEval Estonian Knowledge5.85Knowledge Average Score (%)44.7
EuroEval Finnish Common Sense Reasoning18.32Common Sense Reasoning Average Score (%)43
EuroEval Estonian Common Sense Reasoning5.07Common Sense Reasoning Average Score (%)35.5
EuroEval Estonian NLU28NLU Average Score (%)32.9
EuroEval Estonian20.49Average Score (%)32.1
EuroEval Estonian NLU - MultiWikiQA ET26.75Character F1 (%)27
EuroEval Estonian NLU - EstNER32.99Micro F1 excluding MISC (%)26.5
EuroEval Finnish33.03Average Score (%)21.7
EuroEval Finnish NLU36.71NLU Average Score (%)18.5

Interactive version: theaggregate.ai/model?slug=ahma-2-4b-instruct · How It Works · Data refreshed daily, snapshot 2026-11-08.