Title: Long context %E2%80%94 Benchmarks | LLM EU
Canonical: https://llmeu.com/sv/benchmarks/long-context/
Markdown mirror: https://llmeu.com/sv/benchmarks/long-context.md
Machine-readable: https://llmeu.com/llms.txt (index) - https://llmeu.com/llms-full.txt (full catalogue) - https://llmeu.com/llms-catalogue.json

# Long context

Hur modeller beter sig n%C3%A4r indata %C3%A4r mycket l%C3%A4ngre %C3%A4n en konversation. RULER-raderna m%C3%A4ter h%C3%A4mtning och resonemang inom en l%C3%A5ng indata vid en angiven l%C3%A4ngd; de %C3%A5terst%C3%A5ende raderna rapporterar det maximala kontextf%C3%B6nstret som utgivaren h%C3%A4vdar, vilket %C3%A4r en kapacitetssiffra och inte ett kvalitetsresultat.

Every score in this set is republished from the model publisher%E2%80%99s own model card or vendor report and names that source per entry. LLM EU has run no evaluation, has not reproduced any score, and has not verified the harness settings. Vendor-reported numbers are self-reported and are labelled as such on the model cards. RULER scores come from NVIDIA%E2%80%99s own NeMo Evaluator harness and are not comparable with scores produced by other harnesses. The context-window rows are configuration or vendor claims and say nothing about answer quality at that length.