SpAItial releases Echo-2 HQ with sharper detail and zoom fidelity

• SpAItial shipped Echo-2 HQ as a higher-detail version of Echo-2, and SpAItial_AI's launch post says the model is available both via API and inside the app.
• The official Echo-2 announcement frames Echo-2 as a model that turns text or images into a 3D-consistent world you can explore in real time, while SpAItial_AI's HQ post pitches HQ around sharper detail and closer zoom fidelity.
• SpAItial_AI's example scene thread links straight to shareable worlds, including a favela football field, a jungle stilt house, and a greenhouse artist sanctuary.
• A separate SpAItial_AI retweet ties Echo-2 to a faster downstream workflow: upload a movie still, generate in Echo-2, then export a 3D Gaussian splat PLY file in about 15 minutes.

You can read SpAItial's original Echo-2 launch post, open the linked favela football field scene, and browse the broader Echo-2 gallery. The June 1 update is narrower: SpAItial_AI's launch post is about higher detail, while the workflow retweet points to direct handoff into splat-based 3D scenes.

SpAItial described Echo-2 HQ as an improved model with "greater detail and sharper results." The useful distinction is that this is not a brand new product line. It is a quality-focused update on top of the Echo-2 stack the company introduced in April via its official announcement.

That earlier post says Echo-2 takes either text or image input and returns a 3D-consistent environment rather than a flat video. The June 1 HQ post narrows the claim to what creators will notice first: cleaner surfaces, more appearance fidelity, and better behavior when you zoom in close.

The launch thread pushes people straight into scene links instead of spec-sheet talk. The three examples attached in SpAItial_AI's thread are:

• Favela Football Field Scene
• Jungle Stilt House Retreat
• The Greenhouse Artist's Sanctuary

Those links matter because they show SpAItial treating outputs as explorable worlds, not just rendered hero frames. The homepage makes the same pitch in product language, calling Echo-2 a way to transform a single image into an immersive 3D world that you can view from any angle on spaitial.ai.

The most concrete workflow clue came from the retweeted demo in SpAItial_AI's retweet: "Enter the Backrooms in 15 minutes" by uploading a movie still to Echo-2 and downloading a 3D Gaussian splat PLY file.

For creative teams, that is the interesting part of this small release. HQ is being sold on visual fidelity, but the adjacent post shows where SpAItial thinks the output goes next: into editable, splat-style 3D pipelines built from a single reference image.