GigaPath-Flash and GigaTIME-Flash: Toward population-scale discovery with efficient pathology foundation models
Microsoft says its new Flash pathology models keep near-original performance while cutting compute enough for much larger research cohorts.
GigaPath-Flash uses a 22M-parameter distilled tile encoder and a 21M-parameter slide encoder, scoring within 3% of GigaPath at about 50 times less compute. GigaTIME-Flash is reported as roughly 6 times faster and 8 times lighter on memory than GigaTIME, while matching or improving spatial protein prediction across tested cancer cohorts. Microsoft says both models are open-weight under Apache 2.0, but frames them as research releases, not validated tools for diagnosis or patient-care decisions. Microsoft Research AI's note
GigaPath-Flash uses a 22M-parameter distilled tile encoder and a 21M-parameter slide encoder, scoring within 3% of GigaPath at about 50 times less compute. GigaTIME-Flash is reported as roughly 6 times faster and 8 times lighter on memory than GigaTIME, while matching or improving spatial protein prediction across tested cancer cohorts. Microsoft says both models are open-weight under Apache 2.0, but frames them as research releases, not validated tools for diagnosis or patient-care decisions. Microsoft Research AI's note
score 5