Weights
A complete checkpoint at an exact revision.
We need every required model asset and a stable way to fetch it during packaging and acceptance.
Evidence · Checkpoint or registry path · source revision · asset list
For model teams
Before we integrate a model, we reproduce it.
No deck or application form. These are the same four requirements behind the 10 public runtimes already on Windflow.
See the requirements01 / You provide
Four things before GPU work begins.
These are requirements, not discovery questions. If one is missing, we stop and name it before scheduling compute.
Hosting rights
Permission to store, package, and serve it.
A public license or written authorization must cover hosted inference for Windflow users.
Evidence · License URL or written permission · use restrictions
Runnable path
One clean launch path from input to output.
A Linux GPU environment, container, or launch command must run from a clean setup with one fixed reference input.
Evidence · Repository · setup · launch command · sample input
Measured baseline
A frame-rate number we can reproduce.
Name the hardware, resolution, precision, steps, batch, warmup, sample duration, and whether FPS is native or interpolated.
Evidence · GPU type and count · settings · FPS · method
02 / We build
Once the handoff is complete, the runtime work is ours.
We preserve model-specific behavior and own the production boundary around it.
Model team
Complete handoff
Exact inputs, rights, launch path, and baseline.
Runtime image
Pin source and dependencies, verify weight restore, make startup repeatable, and expose health.
Session contract
Map readiness, lifecycle, reconnects, named tracks, messages, limits, and failures.
Controls + media
Wire only the prompts, images, video, movement, camera, or actions the model really supports.
Recording + usage
Attach recording lifecycle, observability, and ready-time metering to the public session path.
Result
Public session
Measured, documented, recordable, and metered.
Protocol proof / JavaScript SDK
Commands in. A named media track out.
This is the production session surface: a typed command, a reliable control channel, and the generated video on a recvonly track.
- Command
- set_prompt · prompt: string
- Track
- main_video · recvonly · video
import { Windflow } from "@windflow/sdk";
const session = new Windflow.Session({ model: "helios" });
session.on("trackReceived", ({ name, track }) => {
if (name === "main_video") {
video.srcObject = new MediaStream([track]);
}
});
await session.connect(jwt);
await session.sendCommand("set_prompt", {
prompt: "Misty avenue at dawn",
});
await session.sendCommand("start", {});03 / What ships
A runtime, users, and numbers we can defend.
A catalog entry
A production lifecycle label, supported controls, serving profile, and known limitations on the Models page.
Real sessions
The same API and Playground path developers already use across the public Windflow catalog.
Defensible numbers
Windflow-measured FPS, first-frame time, continuity, and control latency attached to a named runtime profile.
Typical production launch
2–4
weeks
011–2 working daysCheck the handoff
023–7 working daysReproduce + package
033–7 working daysIntegrate the session
042–5 working daysJoint acceptance
Custom kernels, multi-GPU topology, license review, or an unstable upstream path pause or extend the clock.
Working terms
Clear before compute starts.
These are the defaults for a public catalog integration. Any exception is agreed in writing before GPU work begins.
No integration fee
Windflow does not charge the model team and covers integration plus public catalog-serving GPU costs. Dedicated or private capacity is scoped separately.
Non-exclusive
Your team keeps its rights and can host, license, or integrate the model elsewhere.
Delist on request
You can request removal at any time. We stop new sessions and agree a wind-down for anything already active.
Review before publish
We review the hardware, settings, method, and measured results together before any performance number goes public.
04 / Direct no
We do not build around a missing foundation.
Announcement only
A paper, teaser, or placeholder repository is not an inference release.
Weights unavailable
We cannot fetch the complete checkpoint and required assets at a stable revision.
Hosting rights unclear
The license does not cover storage and served inference, and there is no written exception.
Baseline does not reproduce
The reference output or FPS depends on missing settings, a private service, manual steps, or an undisclosed interpolation path.
05 / Start the intake
Send the four inputs. Skip the deck.
Email handoffPlain text is enough
- 01Weights and exact revision
- 02Hosting permission or license
- 03Runnable inference path
- 04Reference GPU, settings, and measured FPS