Tupuxuara longicristatus, T. leonardii (Kellner & Campos, 1988, 1994) ~110 mya, Late Aptian, Early Cretaceous this data is based on a museum-grade cast and model of the complete skeleton, but only partial skeletons are known. There is also some confusion with regard to which specimens represent separate species. Derived from a sister to Sinopterus liui, Tupuxuara longicristatus phylogenetically preceded Thalassodromeus and the Goshura Museum cast pecimen attributed to Tupuxuara.

Distinct from Tapejara and Huaxiapterus, the skull and mandible were longer with a dorsally expanded and combined rostral and cranial crest. The mandible had a deeper keel.

The torso was short and deep. The sacrum was coosified.

The sternal complex was large with corners. The humerus was more robust and the deltopectoral crest was smaller. Fingers 1-3 were smaller.

The pelvis was smaller and completely coossified. Judging from the cast, the pes was still digitigrade converging on the pedal configuration of basal Pteranodon. Metatarsal 2 was the longest, but digit 4 extended further from the heel.

The family tree of the Ornithocephalia and Germanodactylia is here. The expanded family tree of the Pterosauria is here. Martill and Naish (2006) among others nest Tupuxuara with azhdarchids, like Quetzalcoatlus, but there is little to justify this placement other than the convergent enlargement of the antorbitral fenestra higher than the orbit.

Javelinadactylus sagebieli (Campos 2021; Latest Cretaceous; TMM 42489-2) seemed to be a good match for Tupuxuara, but an elongate set of cervicals moves this taxon over to Sinopterus.