Viggle released JST-2 on October 19. Instead of learning in pixel or video space, it represents the world as persistent "atoms" in 3D space (Physics-Native Tokens) and uses one architecture to generate scenes, characters and motion. Viggle says a 3D scene takes about one second on an NVIDIA B200 at under $0.01 of inference, with up to 64 input views kept as spatial context, and characters animate without a separate rigging step. It is free to try in the PINOC browser app and is also available via API and MCP.
Key Takeaways
- ✓Scene generation: about 1 second per scene (NVIDIA B200, excluding network latency) at under $0.01 inference cost; Viggle cites about 5–6 minutes for Marble (vendor figures).
- ✓Spatial context: up to 64 input views, expandable one view at a time in real time, output as 3D Gaussian splats rendered locally in the browser.
- ✓Character blind tests (run by Viggle): Average Score 58.4% vs Meshy 7.1 and 73.2% vs Tripo H3.1; face quality 63.0% and 82.0% (50% means parity).
- ✓Motion blind tests (run by Viggle): text-to-motion 74.8% vs Cartwheel Swing and 80.0% vs HY-Motion-1.0; video-to-motion 89.1% vs DeepMotion Animate 3D.
- ✓Access: free in the PINOC browser app, plus API and MCP for agents such as Codex and Claude Code; JST-3 aims to replace rule-based collisions and object interactions with learned physics.
Key Decision Metrics at a Glance
Heavy Claude Code use: compare subscription limits and API bills
Compare 40 dev plans & simulate token costs vs $20/mo subscriptions
Project Links & Resources
Direct AccessIn-Depth Technical Analysis
Background & Industry Pain Points
Most world models learn in pixel or video space, so physical structure has to be inferred from appearance and 3D is reconstructed afterwards. For games and small teams, modeling, rigging and retargeting eat much of the budget. Viggle released JST-2 on October 19 as the second generation of an architecture it has worked on for four years; JST-1 powered its meme video app, which Viggle says has 50M+ users.
Architecture Highlights
JST stands for Joint Space-and-Time. It represents the world as persistent atoms in 3D space, each with an explicit xyz position plus learned vectors for attributes and dynamics, encoded as Physics-Native Tokens. An encoder infers these tokens from real images, a generator produces them from multimodal prompts, and a decoder uses viewpoint plus motion or force inputs to output images or 3D Gaussian splats. Deformation is learned from data, so characters need no separate rigging or skinning.
Authoritative Benchmarks & Measured Results
No third-party benchmarks were published; all figures below are Viggle's own human blind tests (vendor figures, 50% means parity). Character generation Average Score is 58.4% vs Meshy 7.1 and 73.2% vs Tripo H3.1; animation readiness is 73.8% and 88.9%. Text-to-motion scores 74.8% vs Cartwheel Swing and 80.0% vs HY-Motion-1.0; video-to-motion 58.1% vs Cartwheel Comic 4 and 89.1% vs DeepMotion Animate 3D. A scene takes about one second on a B200, versus about 5–6 minutes for Marble per Viggle.
Developer Hands-On Guide
The quickest start is the PINOC browser app with no waitlist: Scene mode generates scenes, and Agent mode combines scenes, characters and motion. Developers can use the API or connect via MCP from Codex, Claude Code and other agents. Motion output works with standard 3D character rigs. Collisions and object interactions still use rule-based physics, and complex clothing and hair are still being improved; JST-3 targets learned physics for these.
Benchmark side-by-side against alternatives, or calculate monthly token cost vs subscription break-even.
Discussion & Comments
0Sign in to join the discussion
Connect with AI developers to exchange benchmark insights.