Spatial concept development
Turn a mood, image, sketch, or written brief into an explorable scene before a full manual 3D build.
Independent tool overview
Echo-2 at a glance
Echo 2 is SpAItial's hosted world-generation model for turning text, images, or 360-degree panoramas into navigable 3D Gaussian Splat scenes. Outputs can be explored in a WebGL browser, edited by region, shared, exported, or generated through an asynchronous API.
Visit the official Echo-2 site ↗
- Inputs
- Text, image, or 360-degree panorama
- Primary output
- Navigable 3D Gaussian Splat world
- Standard generation
- About 8 minutes and 160 credits
- HQ generation
- About 60 minutes and 800 credits
- Credit value
- 1 credit equals $0.01
- Exports
- SPZ, PLY, compressed PLY, and SOG
Overview
What Echo-2 is
Echo 2 creates a persistent spatial scene rather than a fixed video path. The current platform accepts text, standard images, and panoramas; produces browser-viewable Gaussian Splat worlds; and supports region editing, sharing, downloads, and programmatic generation. Standard jobs are listed at roughly eight minutes, while the higher-detail Echo HQ route is roughly one hour.
Its most useful role is rapid spatial visualization: early game worlds, film previsualization, interactive art, virtual staging, and environment prototypes. SpAItial describes the model as physically grounded, but its own current use-case page calls the output appearance-first and places geometry-, semantics-, and physics-oriented representations on the roadmap. Do not treat generated scenes as surveyed reality, BIM, engineering geometry, collision-ready simulation, or a safe robotics environment without downstream validation.
Use cases
Who Echo-2 is best for
The strongest fit depends on the job you need the product to complete, not the size of its feature list.
Film and game previsualization
Scout camera positions, test traversal and mood, and communicate environment direction with a navigable world.
Virtual staging concepts
Explore furniture, material, lighting, and style directions for a room while clearly labeling the result as a visualization.
Interactive web art
Publish lightweight, browser-native Gaussian Splat environments for portfolios, installations, and spatial storytelling.
World-generation integrations
Use the public REST API to create asynchronous text-, image-, or panorama-to-world jobs inside products and pipelines.
Capabilities
Core Echo-2 features
Text-to-world
Generate a navigable environment from a written description.
Image-to-world
Expand a single visual reference into a spatially persistent scene that can be viewed from new positions.
Panorama-to-world
Use a 360-degree panorama as broader visual context for a generated world.
Real-time browser viewer
3D Gaussian Splatting enables interactive WebGL exploration on supported desktop and mobile browsers.
Echo 2 Standard
The default, general-purpose route targets faster generation at 160 credits per world.
Echo HQ
A heavier pipeline targets higher detail and resolution at 800 credits per world.
Region editing
Select part of a world and request changes to objects, colors, materials, lighting, or atmosphere.
Edit drafts and versions
Edits create a draft that can be refined, finalized, discarded, or revisited from a previous version.
World sharing
Share generated environments through public links or more restricted visibility options.
Multiple export formats
Download SPZ, PLY, compressed PLY, or PlayCanvas-oriented SOG assets for downstream use.
Collider mesh export
Paid plans add collider mesh export for downstream interactive workflows.
Asynchronous REST API
Submit generation jobs, poll or receive webhooks, and retrieve signed URLs for splats, panoramas, and thumbnails.
Process
How the Echo-2 workflow works
Step 1
Define the downstream use
Decide whether the output is for concept art, web viewing, staging, previs, a game prototype, research, or a data pipeline.
Step 2
Prepare a rights-cleared input
Use a detailed prompt, clean image, floor plan, or panorama you are authorized to upload, and remove sensitive location data.
Step 3
Generate with Standard first
Use the lower-cost route to test composition and spatial direction before spending five times more credits on Echo HQ.
Step 4
Inspect from difficult viewpoints
Check unseen angles, boundaries, thin objects, mirrors, windows, text, surfaces, scale, occlusion, and camera movement.
Step 5
Edit and export
Refine selected regions, keep useful versions, choose the correct Gaussian format and coordinate convention, then move downstream.
Step 6
Validate for the real application
For architecture, property, games, XR, or robotics, rebuild or verify measurements, collision, physics, semantics, accessibility, performance, and safety in appropriate tools.
Cost
Echo-2 pricing and free plan
SpAItial values one credit at $0.01. Echo 2 Standard costs 160 credits, equivalent to $1.60, and Echo HQ costs 800 credits, equivalent to $8. Credits are shared across the app and API when eligible, while free daily app credits cannot fund API jobs. Check the developer portal because starter-grant and API eligibility language varies across current documentation.
Free
$0
One-time starter balance for trying core creation.
- 1,600 starter credits: up to 10 Standard or 2 HQ generations
- Includes downloads, standard exports, and editing; no collider mesh export
Standard
$24/month
Small-project plan with monthly credits.
- 2,400 credits per month: up to 15 Standard or 3 HQ generations
- Adds collider mesh export
Pro
$48/month
Higher monthly generation allowance.
- 4,800 credits per month: up to 30 Standard or 6 HQ generations
- Includes Standard features with more credits
Custom and Enterprise
Contact SpAItial
Custom volume and support for production workloads.
- Custom generation capacity
- Self-serve top-up credits are also available in custom amounts
Pricing checked . Check current pricing at the source ↗
Assessment
Echo-2 strengths and limitations
Where it stands out
- Creates an explorable world instead of a fixed sequence of video frames
- Supports text, image, and panorama inputs
- Browser-native viewing on supported WebGL 2.0 devices
- Standard generation is priced transparently at 160 credits or $1.60
- HQ route offers a higher-detail option without changing the core workflow
- Region editing and versioned drafts support iterative refinement
- SPZ, PLY, compressed PLY, and SOG exports
- Public asynchronous API with polling and webhook patterns
- Free starter credits provide meaningful evaluation capacity
- Useful across creative, built-environment, and developer prototypes
What to consider
- Vendor-reported state-of-the-art claims and comparisons should be validated independently on the buyer's own use case
- The current output is an appearance-first Gaussian Splat world, not automatically accurate engineering geometry or a clean production mesh
- SpAItial places richer geometry, semantics, dynamics, and physics-oriented representations on the roadmap
- A generated property or factory clone is not a survey, inspection, BIM model, code-compliance review, or safety record
- A visually plausible robotics environment does not establish metric accuracy, collision fidelity, contact dynamics, or safe Sim2Real transfer
- Single-view inputs leave unseen geometry underdetermined, so the model must invent missing structure
- Thin objects, reflections, transparency, repeated patterns, text, boundaries, and occluded regions can break or smear
- Object scale, room dimensions, topology, camera clearance, and walkable surfaces can be wrong
- Gaussian Splats have different editing, lighting, collision, rigging, and engine behavior from conventional polygonal assets
- Collider mesh export on a paid plan does not prove that collision geometry is correct for production
- Echo 2 Standard takes about eight minutes and Echo HQ about one hour before queue variation
- HQ costs five times as many credits as Standard
- World viewing requires WebGL 2.0, and older devices or restricted corporate networks may perform poorly
- VR-headset browsing is not currently supported in the documented compatibility guide
- SPZ exports use an RDF coordinate convention and may need a 180-degree X-axis rotation in third-party viewers
- Free daily app-only credits cannot be used for API requests
- Documentation is inconsistent about whether the one-time free starter balance is directly API-eligible, so developers should verify their key and balance
- Uploaded photographs, floor plans, homes, workplaces, and cultural sites can reveal sensitive locations and private details
- Users need rights to source images, plans, brands, artwork, interiors, people, and generated commercial uses
- Virtual staging and property media should be disclosed clearly so viewers do not mistake visualization for the current physical condition
Compare
Echo-2 alternatives
The right alternative depends on the specific output, workflow, controls and budget your project requires.
Content Creator
World Labs Marble
Compare World Labs Marble for another hosted system focused on persistent generated 3D worlds.
Explore World Labs Marble →Miscellaneous
HY-World 2.0
Consider HY-World 2.0 when an open-source world-model workflow is more important than a managed browser app.
Explore HY-World 2.0 →Design
Blockade Labs Skybox AI
Consider Skybox AI when a 360-degree environment image is sufficient and a navigable generated world is unnecessary.
Explore Blockade Labs Skybox AI →Design
Rodin
Consider Rodin when the goal is an individual polygonal 3D asset rather than an entire Gaussian Splat environment.
Explore Rodin →Questions
Echo-2 FAQs
What is Echo 2?
Echo 2 is SpAItial's hosted world-generation model. It turns text, an image, or a panorama into a navigable 3D Gaussian Splat scene that can be viewed, edited, shared, exported, or generated through an API.
Is Echo 2 a video generator?
No. A video fixes a sequence of frames. Echo 2 produces a persistent spatial representation that a viewer can revisit and explore from different positions.
How much does an Echo 2 generation cost?
Standard uses 160 credits, equivalent to $1.60 at the published one-cent credit value. Echo HQ uses 800 credits, equivalent to $8.
How long does generation take?
SpAItial lists about eight minutes for Standard and about 60 minutes for Echo HQ. Queue position and system load can increase the wait.
What can Echo 2 export?
Current documentation lists SPZ, PLY, compressed PLY, and SOG Gaussian formats. Paid plans also include collider mesh export.
Does Echo 2 create accurate digital twins?
It can create a convincing spatial visualization from sparse input, but it should not be treated as surveyed, metric, engineering, BIM, collision, or physics truth without independent measurement and validation.
Can developers use Echo 2 through an API?
Yes. The public REST API creates asynchronous world jobs from text, files, URLs, or panoramas and returns result URLs after completion.
Can Echo 2 be used for robotics simulation?
It can support appearance and environment prototyping, but safe robotics work requires validated geometry, scale, semantics, collision, physics, domain randomization, and real-world testing beyond the generated scene.
Bottom line
Our Echo-2 verdict
Echo 2 is a strong rapid-prototyping tool when a team wants a navigable, browser-friendly environment instead of a static image or video. Transparent per-generation pricing, a useful free allowance, editing, exports, and an API make it unusually approachable. Its key boundary is representation: current Gaussian Splat worlds are excellent for visual exploration, but they are not automatically production geometry, measured digital twins, or physically valid simulations.
Visit Echo-2 website ↗Similar Content Creator Tools
View all Content Creator AI tools →
Happy Horse
Happy Horse - Alibaba's new SOTA AI video model
Eleven Music
Eleven Music - ElevenLabs' new streaming platform with AI remixing, song generation, and creator payouts
Ernie Image
ERNIE-Image - Baidu's 8B open-weight text-to-image model that nears top rivals on benchmarks despite its small size
ElevenLabs Studio Agents
ElevenLabs Studio Agents - AI co-editor that drafts videos and places sound effects frame by frame
AI training for the future of work.
Get access to all our AI courses, hundreds of real-world AI use cases, live expert-led workshops, an exclusive network of AI early adopters, and more.
AI Courses
Get unlimited access to all of our current & upcoming industry-specific AI courses for the duration of your subscription.
Daily Guides
To keep up with the rapid pace of AI, our team publishes AI implementation guides daily. Our library contains 300+ practical use cases to automate real-world work.
Workshops
Join weekly, live, interactive sessions with industry leaders who are at the forefront of AI for hands-on implementation guidance and exclusive insights.
Community
Network with an exclusive community of AI-first professionals who are working smarter with AI. Learn how early adopters are using AI in their work and businesses.