One model. Both recommendations.
Wan 3.0
WanOne model leads both performance and value in the supplied text-to-video comparison.
Choose your workload
Models for turning written concepts into short video assets.
One model. Both recommendations.
One model leads both performance and value in the supplied text-to-video comparison.
Arena · supplied snapshot·Supplied Arena snapshot · date pending·How we compare
The same model holds both recommendation positions. Its supplied unit price is approximate and actual billing may vary. Arena scores and prices are from the supplied snapshot. The leaderboard source and snapshot date still need confirmation.
The same workload, with the details that help you choose.
| Comparison | Wan 3.0Best Performance & Best Value |
|---|---|
| Arena score | 1476 Supplied Arena snapshot · date pending |
| Relative score | Same recommendation |
| Billing rate | ~$0.07 / second |
| Arena difference | Same model |
| Cost difference | Same model |
| Supported modalities | text → video |
| Main reason to choose | One model holds both recommendation positions. |
| Best fit | Video generation from written concepts |
| Billing details | Approximate supplied reference price per second. Resolution, duration and actual billing may vary. |
The same model holds both recommendation positions. Its supplied unit price is approximate and actual billing may vary.
Implementation
Copy a model test, replace the API key with your own, and run.
Copy the Python or cURL test for Wan 3.0. Replace the API-key placeholder with your own key and run.
Wan 3.0 · text to video · Submits a five-second video test and polls for the resulting video URL.
import json
import time
from urllib.parse import quote
from urllib.error import HTTPError
from urllib.request import Request, urlopen
API_KEY = "YOUR_ROTEN_API_KEY"
def send(url, payload=None):
data = json.dumps(payload).encode("utf-8") if payload is not None else None
request = Request(url, data=data, headers={"Authorization": f"Bearer {API_KEY}", "Content-Type": "application/json"})
try:
with urlopen(request, timeout=180) as response:
return json.load(response)
except HTTPError as error:
raise SystemExit(f"HTTP {error.code}: {error.read().decode('utf-8')}")
payload = json.loads(r'''{
"prompt": "An indigo geometric shape moves slowly across a midnight navy background.",
"duration": 5,
"size": "720p",
"model": "wan3.0-video"
}''')
result = send("https://api.roten.ai/v1/video/generations", payload)
task_id = result["task_id"]
for attempt in range(120):
result = send("https://api.roten.ai/v1/video/generations" + "/" + quote(str(task_id), safe=""))
status = result.get("status")
if status == "completed":
print(result["url"])
break
if status == "failed":
raise SystemExit(json.dumps(result.get("error", result)))
if status not in ("pending", "processing"):
raise SystemExit("Unexpected task response: " + json.dumps(result))
time.sleep(5)
else:
raise SystemExit("The video is still processing. Task ID: " + str(task_id))# Requires curl and Python 3 for reading JSON responses.
set -eu
API_KEY='YOUR_ROTEN_API_KEY'
created=$(curl --fail-with-body --silent --show-error 'https://api.roten.ai/v1/video/generations' \
-H "Authorization: Bearer $API_KEY" \
-H 'Content-Type: application/json' \
--data '{
"prompt": "An indigo geometric shape moves slowly across a midnight navy background.",
"duration": 5,
"size": "720p",
"model": "wan3.0-video"
}')
task_id=$(printf '%s' "$created" | python3 -c 'import json,sys;from urllib.parse import quote;print(quote(str(json.load(sys.stdin)["task_id"]),safe=""))')
attempt=0
while [ "$attempt" -lt 120 ]; do
result=$(curl --fail-with-body --silent --show-error "https://api.roten.ai/v1/video/generations/$task_id" \
-H "Authorization: Bearer $API_KEY")
status=$(printf '%s' "$result" | python3 -c 'import json,sys;print(json.load(sys.stdin).get("status",""))')
case "$status" in
completed) printf '%s' "$result" | python3 -c 'import json,sys;print(json.load(sys.stdin)["url"])'; exit 0 ;;
failed) printf '%s\n' "$result" >&2; exit 1 ;;
pending|processing) sleep 5 ;;
*) printf '%s\n' "$result" >&2; exit 1 ;;
esac
attempt=$((attempt + 1))
done
printf 'Video still processing. Task ID: %s\n' "$task_id" >&2
exit 1Python tests use the standard library. Video cURL tests also use Python 3 to read task responses.
A clear reference point
Recommendations combine the supplied Arena scores with the supplied model prices. A benchmark score is one signal for choosing a model for your workload.
Arena · supplied snapshot·Supplied Arena snapshot · date pending·How we compare
Built for production
Talk to us about volume pricing, model selection and migration.