Head Swap – Generate
namifusion/head-swap/generate
Replace selected heads in the video with your reference images using Wan2.2-Animate. Keeps the original audio track.
Examples
Parameters
| Name | Type | Default | Constraints | Description |
|---|---|---|---|---|
| track_manifest_url *Track manifest URL | text | — | — | track_manifest_url returned by the analyze model |
| mappings *Head mappings | array<object> | — | — | Array of {obj_id, img_urls}. obj_id comes from analyze output; 1-3 reference image URLs per head. |
| ↳obj_id *Head obj_id | number | — | — | Object identifier returned by the corresponding analyze request. |
| ↳img_urls *Reference images | image_upload | — | 1–3 items | Reference image URLs associated with this detected object. |
| resolutionResolution | select | 720p | 480p | 720p | Selects the output resolution or provider generation mode. Higher settings may take longer to process. |
| seedSeed | number | — | — | Optional random seed for reproducible results; leave empty to use a random seed. |
Output fields
| Field | Type | Description |
|---|---|---|
| videos | array | Result video URL(s), original audio track preserved |
API
Call this model through one unified REST API. Get a key on the API Keys page.
cURL
# 1) Submit — returns { "task_uuid": "..." }
curl -X POST "https://www.namifusion.com/api/v1/marketplace/run/namifusion/head-swap/generate" \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"input": {
"track_manifest_url": "ANALYZE_OUTPUT.track_manifest_url",
"mappings": [
{
"obj_id": 0,
"img_urls": [
"https://assets-public.namifusion.com/marketplace/images/2026-06-12/638e1ee18ac7.jpeg"
]
}
],
"resolution": "720p"
}
}'
# 2) Poll until status is "completed", then read the output URLs
curl "https://www.namifusion.com/api/v1/marketplace/run/tasks/TASK_UUID" \
-H "Authorization: Bearer YOUR_API_KEY"Python
import time, requests
API_KEY = "YOUR_API_KEY"
HEADERS = {"Authorization": f"Bearer {API_KEY}"}
# 1) Submit
resp = requests.post(
"https://www.namifusion.com/api/v1/marketplace/run/namifusion/head-swap/generate",
headers=HEADERS,
json={
"input": {
"track_manifest_url": "ANALYZE_OUTPUT.track_manifest_url",
"mappings": [
{
"obj_id": 0,
"img_urls": [
"https://assets-public.namifusion.com/marketplace/images/2026-06-12/638e1ee18ac7.jpeg"
]
}
],
"resolution": "720p"
}
},
)
resp.raise_for_status() # 401/402/429/5xx stop here instead of polling a bad task
task = resp.json()
# 2) Poll until a terminal state (completed / failed / cancelled).
# This model is allowed up to 7200s server-side.
deadline = time.time() + 7260
while task.get("status") not in ("completed", "failed", "cancelled"):
if time.time() > deadline:
raise TimeoutError(f"still {task.get('status')} — keep the task_uuid and poll later")
time.sleep(3)
poll = requests.get(f"https://www.namifusion.com/api/v1/marketplace/run/tasks/{task['task_uuid']}", headers=HEADERS)
poll.raise_for_status()
task = poll.json()
print(task["status"], task.get("output"))JavaScript
const API_KEY = "YOUR_API_KEY";
const HEADERS = { Authorization: `Bearer ${API_KEY}` };
// 1) Submit
const resp = await fetch("https://www.namifusion.com/api/v1/marketplace/run/namifusion/head-swap/generate", {
method: "POST",
headers: { ...HEADERS, "Content-Type": "application/json" },
body: JSON.stringify({
"input": {
"track_manifest_url": "ANALYZE_OUTPUT.track_manifest_url",
"mappings": [
{
"obj_id": 0,
"img_urls": [
"https://assets-public.namifusion.com/marketplace/images/2026-06-12/638e1ee18ac7.jpeg"
]
}
],
"resolution": "720p"
}
}),
});
if (!resp.ok) throw new Error(`submit failed: ${resp.status} ${await resp.text()}`);
let task = await resp.json();
// 2) Poll until a terminal state (completed / failed / cancelled).
// This model is allowed up to 7200s server-side.
const deadline = Date.now() + 7260 * 1000;
while (!["completed", "failed", "cancelled"].includes(task.status)) {
if (Date.now() > deadline) throw new Error(`still ${task.status} — keep the task_uuid and poll later`);
await new Promise((r) => setTimeout(r, 3000));
const poll = await fetch(`https://www.namifusion.com/api/v1/marketplace/run/tasks/${task.task_uuid}`, { headers: HEADERS });
if (!poll.ok) throw new Error(`poll failed: ${poll.status}`);
task = await poll.json();
}
console.log(task.status, task.output);Documentation
Head Swap - Generate
Replace selected heads in an analyzed video with reference images while preserving the original audio.
Head Swap Generate is the second stage of the Head Swap workflow. Each item in mappings associates an Analyze obj_id with one to three reference image URLs, making the replacement target explicit. The video_url and track_manifest_url must come from the same Analyze task, and the generation resolution must match the analysis resolution.
Key Features
- Explicit per-head selection: Uses the
obj_idfrom Analyze to avoid replacing the wrong target. - Multiple-target mappings: Assign separate reference images to several heads in one video.
- Multiple reference images: Supply one to three images per target to cover appearance and viewing-angle details.
- Original audio preservation: Keeps the input video's audio track in the completed output.
- 480p and 720p output: Select the quality tier that matches the intended result and existing pricing band.
Technical Specifications
| Item | Details |
|---|---|
| Workflow | Head Swap step 2 |
| Required inputs | video_url, track_manifest_url, mappings |
mappings format | Array of {obj_id, img_urls} objects |
| Reference images | 1-3 URLs per target |
| Resolution | 480p, 720p (default) |
| Maximum video length | 30 seconds |
| Optional input | seed |
| Output | Video URL with the original audio preserved |
Basic Workflow
- Process the input video with Head Swap Analyze.
- Select the target
obj_idvalues from the previews inobjects. - Pair each ID with its head reference image URLs in
mappings. - Submit the same video and returned Analyze manifest, then poll until generation completes.
Pricing
The existing Marketplace rates are 40 credits per 5-second unit at 480p and 75 credits per 5-second unit at 720p. Videos are limited to 30 seconds. This documentation update does not change pricing.
Best Practices
Use sharp references that show the full head, ideally with camera angles and lighting close to the target video. When supplying multiple images, keep the person's appearance consistent across them. The video, manifest, and obj_id values must come from the same Analyze result; never mix values from separate tasks.
Related models
Head Swap – Analyze
Detect and track heads in a video with SAM3. Returns a head list with preview images and a track manifest for the generate step.
Character Swap – Generate
Replace selected characters in the video with your reference images using Wan2.2-Animate. Keeps the original audio track.
Character Swap – Analyze
Detect and track every person in a video with SAM3. Returns a character list with preview images and a track manifest for the generate step.