Wan 2.5 is an advanced AI model for generating high-quality, photorealistic videos directly from text prompts with synchronized audio. It represents a major leap in video generation technology, featuring native 4K support, cinematic controls, and natural motion synthesis. Designed for creators seeking professional-grade storytelling and emotional fidelity, Wan 2.5 delivers immersive, multi-minute video clips with fluid motion and precise audio-visual synchronization.
vs Google Veo 3: Wan 2.5 supports native 4K video, longer clips, multilingual audio-visual synchronization including Chinese, and dynamic cinematic camera controls, while Veo 3 is limited to 1080p, shorter clips, English-centric audio sync, and basic fixed shots. Wan 2.5 is also more cost-efficient for creators with full audio input support versus Veo 3's system-generated sound only.
vs Runway Gen-4: Wan 2.5 excels in efficient real-time audio-video sync and native 4K output, offering enhanced motion fidelity and flexible camera workflows, whereas Runway Gen-4 focuses more on post-production effects and in-browser editing features but lacks deeper audio integration.
vs Pika Labs: Wan 2.5 generates longer continuous narrative videos with fine-tuned cinematic controls and multilingual voice syncing, compared to Pika Labs which provides faster short clip generation mainly for social media formats without advanced camera or audio sync features.
vs Kling 2.5 Turbo: Wan 2.5 offers superior photorealistic character rendering and precise lip-sync across languages, plus multiple video size outputs, while Kling 2.5 Turbo is optimized for high-speed generation and stylized animation effects but with less audio-visual integration.
Accessible via AI/ML API. Documentation: available here.
import requests, time headers = {"Authorization": "Bearer " + AIMLAPI_KEY} job = requests.post( "https://api.aimlapi.com/v2/video/generations", headers=headers, json={ "model": "alibaba/wan2.5-t2v-preview", "prompt": "A serene timelapse of clouds over a mountain range" }, ).json() gid = job["id"] while True: res = requests.get(f"https://api.aimlapi.com/v2/video/generations?generation_id={gid}", headers=headers).json() if res.get("status") in ("completed", "error"): break time.sleep(5) print(res["video"]["url"])
const headers = { Authorization: `Bearer ${process.env.AIMLAPI_KEY}`, "Content-Type": "application/json", }; const job = await (await fetch("https://api.aimlapi.com/v2/video/generations", { method: "POST", headers, body: JSON.stringify({ "model": "alibaba/wan2.5-t2v-preview", "prompt": "A serene timelapse of clouds over a mountain range" }), })).json(); let res; do { await new Promise((r) => setTimeout(r, 5000)); res = await (await fetch(`https://api.aimlapi.com/v2/video/generations?generation_id=${job.id}`, { headers })).json(); } while (!["completed", "error"].includes(res.status)); console.log(res.video.url);
# submit the generation — the response contains the job "id" curl -X POST https://api.aimlapi.com/v2/video/generations \ -H "Authorization: Bearer $AIMLAPI_KEY" \ -H "Content-Type: application/json" \ -d '{"model":"alibaba/wan2.5-t2v-preview","prompt":"A serene timelapse of clouds over a mountain range"}' # then poll until status is "completed" — video.url holds the result curl "https://api.aimlapi.com/v2/video/generations?generation_id={id}" -H "Authorization: Bearer $AIMLAPI_KEY"
OpenAI-compatible — swap the base URL and it works with your existing SDK.
| Type | Price |
|---|---|
| Output | |
| Model | Input | Output | Context | Best for |
|---|---|---|---|---|
Wan 2.5 Text-to-Video Preview This page | Video generation |
Wan 2.5 Text-to-Video Preview takes text as input and returns video.
Wan 2.5 Text-to-Video Preview became available on November 20, 2025.
Wan 2.5 Text-to-Video Preview is priced at $0.065 / sec (variable).
Wan 2.5 Text-to-Video Preview is billed per generation — a fixed charge per output rather than by prompt length.
Wan 2.5 Text-to-Video Preview was built by Alibaba Cloud.
Send a request to v2/video/generations with alibaba/wan2.5-t2v-preview as the model id.
Yes. Wan 2.5 Text-to-Video Preview is served through AI/ML API, so the same key and endpoint format used for other models applies.
It can generate video in 480p, 720p, and 1080p resolutions.
-min.webp){kind=small}
