← Models
Navigable world model
LingBot-World 2.0 API
A navigable stream on main_video you steer in real time. Served on Windflow as a realtime session you can plug into your product and pay for by the second.
Windflow session capture
Overview
What LingBot-World 2.0 does
LingBot-World is a world-model family from Robbyant, Ant Group's embodied-intelligence unit. On Windflow, the LingBot-World 2.0 profile takes a seed image and a text prompt and returns a generated stream steered with persistent movement and look controls. Check the exact version's weight terms separately from its code license.
A session keeps the model running: connect once, steer it with text prompts, a reference image and movement and look controls while it runs, and receive the live stream back on a named output track.
Every model in the catalog speaks the same typed session protocol, so switching between models changes the controls you send, not how you integrate.
- Architecture
- Seed-image conditioned navigable world model
- Resolution
- 1408 × 768 delivery (704 × 384 generation)
- Chunk
- Movement and look apply at the next generation-block boundary
- Inputs
- A seed image (uploadFile) and a text prompt, plus persistent WASD movement and look controls
- Outputs
- A navigable stream on main_video you steer in real time
- Delivery
- WebSocket (H.264) · WebRTC
- Session protocol
- sdk.realtime.v2
- Price
- $0.0024 per billable second
$0.0024
Per billable second
Meter starts when generation begins
$8.64
Per billable hour
The same rate, held for an hour
Production
Runtime status
Capacity is brought up on demand
Private beta
Access
Request access or a live demo
Use cases
What teams build with LingBot-World 2.0
- Navigable game-like worlds generated from a single image.
- Interactive scene exploration for previsualisation and level concepting.
- Embodied-AI and simulation research that needs a controllable world stream.
- Prototyping WASD-controlled experiences without building a game engine first.
Provenance
Where LingBot-World 2.0 comes from
The model and its code are published by the upstream authors. Windflow runs it behind one session API and bills per billable second.
- Robbyant/lingbot-world on GitHubLicence · Apache-2.0 (repository)
Compare
LingBot-World 2.0 next to similar models
| Model | Inputs | Output | Price per billable second |
|---|---|---|---|
| LingBot-World 2.0 | A seed image (uploadFile) and a text prompt, plus persistent WASD movement and look controls | A navigable stream on main_video you steer in real time | $0.0024 |
| DreamX-World-5B | A seed image, a text prompt, and movement/camera-control state | A controllable world rollout on main_video | $0.0023 |
| ABot-World-0-5B | A seed image, a text prompt, and persistent movement/camera key state | A third-person navigable world stream on main_video | $0.0023 |
| Hy-Nomas | Text and persistent camera movement/look snapshots | Silent generated video on main_video | $0.0047 |
FAQ
Questions about LingBot-World 2.0
- How do I control LingBot-World?
- With persistent movement and look state: set a direction and the world keeps moving until you send idle. A text prompt and event triggers shape what the world contains.
- Can I host LingBot-World in production on Windflow?
- Windflow offers the LingBot-World 2.0 profile as a hosted realtime session in private beta. Billing starts when generation starts, with the published grace policy. Request access and evaluate controls, continuity, licensing and current readiness before committing production traffic.
- How do I prevent movement continuing after a player releases a key?
- Send the complete held-key snapshot with a rising input_seq, and send an empty snapshot on key release where appropriate and on window blur. Stale snapshots are acknowledged but not applied. Persistent movement continues until replaced; an acknowledgement is not the later visible response.
- Can I use the generated world as a deterministic game engine?
- Do not assume generated video provides stable geometry, collision rules or authoritative game state. Keep those rules in your application and evaluate a repeatable route for visual drift, response delay and control limits.
- How much does LingBot-World 2.0 cost on Windflow?
- LingBot-World 2.0 costs $0.0024 per billable session-second ($8.64/hr of billable time). Billing starts at the session's first generation_started event and stops when the session is released or ends. Queueing, worker startup, time before generation starts and the documented reconnect grace are not billed. After generation starts, paused or unwatched time remains billable; release the owner session to stop the meter. There is no subscription.
- How do I get access to LingBot-World 2.0?
- Windflow is in private beta. Request access from the sign-up page, or ask for a live demo of LingBot-World 2.0. Approved accounts start with $5 and four concurrent sessions.
- Is LingBot-World 2.0 always running?
- No. Model capacity is brought up on demand rather than kept standing. Live readiness for each model is published at api.windflow.online/v1/status, and a live demo can be arranged on request.
Guides and comparisons
- Live audience use caseBuild a shared AI video live stream for an audience
- GuideChoosing a realtime video model: a decision guide for 12 hosted models
- GuideHow per-second pricing works for realtime video, with worked examples
- GuideHow to add live AI video to your app and pay per second
- Evaluation guideHow to measure realtime video API latency and responsiveness
- Model API guideHow to use the LingBot API
- RadarRealtime models radar: week of October 5, 2026
- Cost comparisonRealtime video API pricing: compare rates and billable timelines
- GuideStreaming a video diffusion model to the browser: WebRTC or WebSocket?
- Endpoint comparisonWindflow vs fal: choosing the right realtime video endpoint
- Platform comparisonWindflow vs Reactor: choosing a realtime model session API
- World model guideWorld model API: embed a navigable AI world in your app
- Use caseNavigable world models for games: an API with WASD control
- GuideWhat is a world model? How it differs from video generation
- Troubleshooting guideHow to recover a realtime video stream after a dropped connection
Other models
Realtime video generationHeliosRealtime video generationLongLive-2.0Live video transformationSANA-StreamingLive video transformationEdenLiveLive video transformationJoyAI-Video-EditNeural game renderingMatrix-Game 4.0Navigable world modelDreamX-World-5BNavigable world modelABot-World-0-5BRealtime video generationMiniMax FastH3Realtime video generationLTX-2.5Navigable world modelHy-Nomas