all the models — AI benchmark observatory
← Benchmarks

NoLiMa 32K

NoLiMa evaluated at a 32768-token context length. Tests latent associative reasoning in long contexts with minimal lexical overlap between questions and needles.

id nolima-32k · max 1 · 1 models reported

#ModelScore

No scores for this benchmark yet.