The satellite is now being asked to file its own understanding
the satellite is now being asked to file its own understanding: nasa and ibm open source prithvi became the first geospatial foundation model run in orbit, a zero shot vision language model classified scenes onboard a small spacecraft in april, and a twenty one million parameter student distilled from a billion parameter earth model beat rivals some orders of magnitude larger.
the orbit now files its own inference log.
Context
NASA's science site, in a post of May 7, 2026, says researchers from Adelaide University, ESA Phi-lab, Thales Alenia Space and the SmartSat Cooperative Research Centre uploaded and demonstrated NASA and IBM's open-source Prithvi Geospatial foundation model aboard two in-orbit platforms, making it the first geospatial foundation model deployed in orbit. NAVI-Orbital, in an arXiv paper of June 2026, says that on April 16, 2026 its system on a Low Earth Orbit spacecraft achieved what the authors believe is the first in-orbit demonstration of a vision-language model performing autonomous multi-modal inference entirely onboard, using Gemma 3 to classify each captured scene, with 88.16% accuracy on a 7,960-image ground benchmark and no fine-tuning for the flight imager. TESSERA v2, an arXiv paper of July 2026, reports a 21-million-parameter student distilled from a 1B-parameter teacher that the authors say outperforms all open and proprietary models tested, some orders of magnitude larger.
These are three separate projects, and the post's phrasing groups them by theme. The first in orbit claims are NASA's for Prithvi and the authors' own to their knowledge for NAVI-Orbital, so neither is independently verified. The NAVI-Orbital accuracy is a ground benchmark figure and not an in-orbit score. The TESSERA claim is the authors' own aggregate over their 29-task suite, and the distilled model is an Earth observation embedding model that was not run in orbit in the sources read. The orbit now files its own inference log is the author's line.
Watch next
- Independent evaluation of the in-orbit demonstrations.
Sources
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 4 October 2026 at 21:40 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →