gpt-sw3-1.3B-instruct — benchmark results

Provider: AI Sweden. Released 2023-04-28. Access: Open.

Unified ELO 1241 ± 89, rank #1639 of 1776 rated models, from 18 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
EuroEval Swedish NLU - Swerec73.34Sentiment classification Score (%)42.5
EuroEval Norwegian NLU - Norec35.58Sentiment classification Score (%)36.8
EuroEval Danish NLU - Angry Tweets27.14Sentiment classification Score (%)24.6
EuroEval Danish15.44Average Score (%)16.8
EuroEval Norwegian8.88Average Score (%)16.6
EuroEval Danish NLU - Multi Wiki QA DA17.3Reading comprehension Score (%)15.2
EuroEval Swedish NLU - Multi Wiki QA SV15.01Reading comprehension Score (%)14.2
Icelandic LLM - WinoGrande-IS52.11Score (%)13.7
EuroEval Norwegian Knowledge1.29Knowledge Average Score (%)13.3
Icelandic LLM - GED50.5Score (%)13.2
Icelandic LLM - WikiQA-IS2.84Score (%)11
Icelandic LLM - Belebele-IS22.89Score (%)8.8

Interactive version: theaggregate.ai/model?slug=gpt-sw3-1-3b-instruct · How the rankings work · Data refreshed daily, snapshot 2026-07-22.