VQV Signal
According to a post on X by @speridlabs, Sperid Labs released Iris-Z in 9B and 32B sizes as one backbone that generates images with depth, estimates depth and conditions on it. The lab says Iris-Z beats task-specific depth models such as MoGe-3. It also says the weights, code and a demo are released under Apache 2.0. The post is mainly a video, and these performance claims come from the lab, not from independent testing that VQV has checked.
According to a post on X by @speridlabs, Sperid Labs released Iris-Z in 9B and 32B sizes as one backbone that generates images with depth, estimates depth and conditions on it. The lab says Iris-Z beats task-specific depth models such as MoGe-3. It also says the weights, code and a demo are released under Apache 2.0. The post is mainly a video, and these performance claims come from the lab, not from independent testing that VQV has checked.
AI-assisted summary based on the listed source.
Sperid Labs released Iris-Z in 9B and 32B sizes, one backbone that generates images with depth, estimates depth and conditions on it. The lab says it beats task-specific depth models such as MoGe-3 and released weights, code and a demo under Apache 2.0; the post is mainly a video. Why it matters: Open weights that...
Builders in 3D, robotics and graphics would get one open model for both appearance and geometry, rather than separate tools for each. Because the weights are released under Apache 2.0, teams can test and build on the model directly. Independent benchmarks against MoGe-3 and other depth models will show whether the lab's claims hold up.
VQV organizes public signals from inspectable sources. It does not independently verify the underlying report.
Signal Strength reflects source quality, relevance, freshness and evidence. Public Interest helps organize discovery; it is not proof of truth.
VQV surfaced this signal because it is recent, relevant to Open Source LLMs, connected to Scobleizer News Pulse (X).
No login, cookies, social SDKs, or automatic posting.
