Locamage (@Locamage) profile picture

Locamage

@Locamage

Understanding the Earth

locamage.com
Joined October 2026
  • Tweets14
  • Following88
  • Followers315
4Media
Photo by Locamage (@Locamage)Photo by Locamage (@Locamage)Photo by Locamage (@Locamage)Photo by Locamage (@Locamage)
Meet Finch 1, our first step toward one model that works with any sensor, at any resolution, anywhere on Earth. Trained on imagery from 50+ sensors across satellites, aircraft and drones.
32
32
9
406
38,020
Day 3 of Finch 1: what changed. Crop types in France through a whole season, and crop mapping from radar alone. Same frozen model, same single linear layer.
Image from Locamage (@Locamage)
Image from Locamage (@Locamage)
Locamage (@Locamage) profile pictureLocamage@LocamageOct 27
Day 2 of Finch 1: what it sees. Land cover in France, buildings in Khartoum, roads and roofs around the Chesapeake Bay, and a wildfire scar in the US, from 20 cm aerial photos to 30 m satellite imagery. Every map comes from one frozen model and a single linear layer.
Image from Locamage (@Locamage)
Image from Locamage (@Locamage)
Image from Locamage (@Locamage)
Image from Locamage (@Locamage)
4
1
12
223
On public benchmark test sets: • Crops (PASTIS): +11.3 vs AlphaEarth • Radar crop mapping (CropHarvest China): +8.4 vs OlmoEarth v1.3 Step through the season: locamage.com/blog/what-chang…
27
Day 2 of Finch 1: what it sees. Land cover in France, buildings in Khartoum, roads and roofs around the Chesapeake Bay, and a wildfire scar in the US, from 20 cm aerial photos to 30 m satellite imagery. Every map comes from one frozen model and a single linear layer.
Image from Locamage (@Locamage)
Image from Locamage (@Locamage)
Image from Locamage (@Locamage)
Image from Locamage (@Locamage)
Locamage (@Locamage) profile pictureLocamage@LocamageOct 26
Meet Finch 1, our first step toward one model that works with any sensor, at any resolution, anywhere on Earth. Trained on imagery from 50+ sensors across satellites, aircraft and drones.
1
3
1
11
585
Every example comes from a public benchmark: FLAIR2, SpaceNet2, m-Chesapeake and HLS Burn Scars. See them all, with a slider between imagery and map: locamage.com/blog/what-finch…
What Finch sees

What Finch sees

Segmentation maps from four public benchmarks, from 20 cm aerial photos to 30 m satellite imagery, all read by one frozen model.

locamage.com
2
43
Full results and how we compared: locamage.com/blog/finch-1 Request early access: locamage.com/#access
Finch 1: one model for every sensor

Finch 1: one model for every sensor

Finch 1 is our first geospatial foundation model, trained on imagery from more than 50 sensors across satellites, aircraft and drones.

locamage.com
7
655
And it keeps getting better as we scale. Full results and how we compared: locamage.com/blog/finch-1
Finch 1: one model for every sensor

Finch 1: one model for every sensor

Finch 1 is our first geospatial foundation model, trained on imagery from more than 50 sensors across satellites, aircraft and drones.

locamage.com
1
15
1,182
If you are a researcher and want to work with us, email [email protected]. Thanks to a16z Speedrun Alpha (@speedrun) and Google's TPU Research Cloud and TPU Builder Program for supporting this work.
17
964
12 hours
7
301
Finch
1
2
5
318
Load more