Seed GR-3

Name: Seed GR-3
Author: ByteDance Seed

other

A robust vision-language-action model designed to be generalizable and capable of executing long-horizon and dexterous tasks.

Capabilities

Show unsupported (8)

No supported capabilities recorded.

Model IDseed-gr-3

Creatorbytedance

Family—

License—

Statusactive

Typeother

Input modalitiesimage

Output modalities—

Open weight—

Parameters—

Architecture—

Context window—

Max input—

Max output—

Reasoning tokens—

Knowledge cutoff—

Release date—

Sourceofficial

GET/v1/models/bytedance/seed-gr-3

seed-gr-3

A robust vision-language-action model designed to be generalizable and capable of executing long-horizon and dexterous tasks.

Changes · 2 entries

seed-gr-3createca7a400Nov 2, 2026, 03:02 AM

seed-gr-3create08f0ffeMay 11, 2026, 08:27 AM