Image generationQwen + Wan. One API.Ship faster with BasicRouter.
- Turn a product photo into a 5-second video ad, ready to test before lunch Video ads
- Generate posters where the text is actually readable — no designer required Readable posters
- Drop in a screenshot and get structured data back in one API call Structured data
- $5 free credit gets you 166 images or 50 seconds of video, no card needed $5 free
Trusted by teams running on
What each model brings to the table.
QWEN3.6-PLUS1M context. Text, images, and video in one call.
The strongest cost-to-performance ratio in the Qwen family. Feed it a mix of text, screenshots, and video frames — it reasons across all of them.
- Upload a screenshot and get working HTML/CSS back
- Debug a 200-file repo in a single session
- Summarize a 500-page document with key findings
QWEN3-VLIt sees what's on screen and understands it.
Point it at a screenshot, a dashboard, a scanned document, or a messy PDF. It reads the layout, extracts fields, and answers follow-up questions.
- Dashboard capture to anomaly and root cause
- Invoice scan to every field as structured JSON
- Product photo to a detailed catalog description
QWEN-IMAGE + WAN2.7AI-generated images where the text is actually readable.
Qwen-Image renders multi-line text that other models can't — titles, taglines, body copy. Wan2.7 adds photorealism and style transfer on top.
- A movie poster with title, cast, and tagline you can read
- Product hero image edits without opening Photoshop
- Twenty ad variations in under a minute to find the winner
WAN VIDEOFrom a text prompt to a finished clip.
Describe a scene and get a video. Upload a still and watch it move. Or re-edit existing footage with a single instruction. Supports 720p and 1080p.
- Product still to a 5-second rotating teaser
- Text prompt to a cinematic clip with camera movement
- Existing footage restyled or re-shot, no crew needed
How teams actually use it.
E-commerce
Generate 50 product photos and ad variants in one morning — no photographer, no studio.
qwen-image-2.0, wan2.7-image
“We cut product photography costs by 80%. What used to take two weeks now ships same day.”
— Sarah L., E-commerce Manager80% lower photo costs. Same-day turnaround.
Marketing
Localize a campaign across 15 markets — swap text, adjust visuals, generate video cuts — in a single afternoon.
qwen-image-max, wan2.6-t2v
“Our team of two now ships 30 short video ads a week. Before this, we could barely do five.”
— James K., Growth Marketer30 short video ads a week. 6x output with a team of two.
Developer
Drop an entire repo into the 1M context window. Ask it to find bugs, write tests, or generate documentation.
qwen3.6-plus, qwen3-max
“I dumped our entire repo in and asked it to find the auth bug. It found it in 12 seconds.”
— Wei C., Senior EngineerAuth bug found in 12 seconds.
Finance
Photograph 200 invoices, get structured JSON back — vendor, line items, totals, tax, due dates.
qwen3-vl-flash, qwen3-vl-plus
“500 invoices in 20 minutes. Our monthly close moved up by three days.”
— Maria R., Finance Ops Lead500 invoices in 20 minutes. Monthly close 3 days faster.
Content Creator
Turn a selfie into a 3-second animated intro for Reels. Generate 10 thumbnail concepts in one click.
wan2.6-i2v-flash, qwen-image-2.0-pro
“Every Reel gets a polished animated intro now. My engagement went up 25% in the first month.”
— Alex T., Creator25% higher engagement in month one.
Designer
Go from client brief to 40 visual concepts in one hour. Apply a brand's illustration style to 50 new assets via API.
qwen-image-max, qwen-image-2.0-pro
“I showed the client 40 concepts in the first meeting. They signed on the spot.”
— Priya M., Brand Designer40 concepts in the first meeting. Client signed immediately.
Real Estate
Photograph an empty room. Get it virtually staged in three styles — modern, mid-century, family — at $0.03 per image.
wan2.7-image-pro, qwen-image-plus
“Virtual staging at a dollar a room instead of three thousand. Listings move faster too.”
— David H., Listing Agent$1 virtual staging per room instead of $3,000.
Customer Support
Customer sends a screenshot of their error. The model reads it, identifies the issue, and drafts a response with exact steps.
qwen3-vl-flash, qwen3.5-plus
“First-response time went from eight minutes to ninety seconds. We handle triple the volume.”
— Lisa N., Support LeadFirst response in 90 seconds. 3x ticket volume per shift.
Researcher
Upload 20 papers into the 1M context window. Ask which ones contradict each other, or draft a literature review.
qwen3.6-plus
“A literature review that used to take me two weeks now takes an afternoon of back-and-forth.”
— Dr. Tom Y., Research ScientistTwo weeks of reading compressed into one afternoon.
Small Business
Describe your business. Get 20 logo concepts, a social media content calendar for the month, and 4 short video ads.
qwen-image-max, qwen3.5-plus, wan2.6-t2v
“I built our entire brand identity and a month of social content in one Sunday. Under ten dollars.”
— Riku S., FounderFull brand plus 30 days of content for under $10.
See it in action.
MonsterMMORPGon Hugging Face
Never seen such a good public model. This model is the FLUX impact of video generation models.
ptwonliner/StableDiffusion on Reddit
wan2.6-t2v demonstrates significantly improved understanding of complex prompts.
Useful_Ad_52r/StableDiffusion on Reddit
Just in case you didn't free up some space, be ready for 10 sec 1080p generations.
wan2.6-t2v
010203040506
exportfunctionHeroPanel() {
return (
<sectionclass="campaign-grid">
<h1>Readable AI visuals</h1>
<button>Claim your $5 credit</button>
</section>
);
}
Why BasicRouter.
We're an infrastructure partner of Alibaba Cloud, AWS, Google Cloud, and Azure. Over 150 enterprises run production workloads on our network.
Up to 5xcheaper on select models
150+enterprises in production
$0.03per image, starting from
- Built on Alibaba Cloud — same backbone behind enterprise AI deployments worldwide
- One OpenAI-compatible endpoint — drop it into your existing stack with a single line of code
- Every Qwen and Wan model under one API key, no per-model authentication
- What you see is what you pay — no hidden markup, no surprise fees
Transparent pricing. No surprises.
Most Popular
Pay As You Go
From $0.03/image
- Qwen3.6-Plus: $0.35 / 1M tokens
- Qwen3-VL-Plus: $0.20 / 1M tokens
- Qwen-Image-2.0: $0.03 / image
- Wan2.6-T2V: $0.50 / video
- Wan2.7-Image: $0.04 / image
For Scale
Volume Pricing
Custom rates
- Dedicated account management
- Custom SLA agreements
- Volume discounts up to 40%
- Priority support
- Custom model deployment
- On-premise options available
For Teams
Team Plan
From $99/month
- Shared credit pool for team
- Team member management
- Usage analytics dashboard
- Priority email support
What you can build for under $5
Test 20 directions, make a product video, or analyze a 500-page report — all within your free credit.
Test 20 ad directions
Generate 20 poster variations with wan2.7-image in under two minutes, then move forward with only strongest concepts.
Make a product teaser
Turn a single product photo into a 5-second rotating video for about $0.50.
Analyze a long report
Hand a 500-page document to qwen3.6-plus and get back a summary with key findings and structured data.
Turn a screenshot into code
Upload a design screenshot and let Qwen3.6-Plus write HTML, CSS, and responsive layout for you.
Process 200 invoices
Upload photos of invoices and get structured JSON back — vendor, line items, totals, tax, due dates.
Localize a campaign
Generate 15 localized versions of a poster or video ad — swap text, adjust visuals, in one afternoon.
Common questions.
What does $5 of credit get me?
Approximately 166 images with wan2.7-image, 50 seconds of 720p video with wan2.6-t2v, or thousands of document analysis calls with qwen3-vl-flash. Enough to validate an end-to-end use case before committing real budget.
How do your prices compare to other resellers?
We're an Alibaba Cloud infrastructure partner with direct access to model serving. On most models, our prices match official rates. On select models like qwen3.6-plus, we're up to 5x cheaper than resellers who mark up the same weights.
Which model should I try first?
qwen3.6-plus for general work — it has a 1M-token context window and handles text, images, and video. qwen3-vl-flash for affordable visual understanding. wan2.7-image for image generation at $0.03 per image.
When do I use Qwen-Image versus Wan2.7 Image?
Qwen-Image when you need readable text in the output — posters, infographics, layouts. Wan2.7 Image for photorealistic quality, style transfer, and character consistency. Wan Video for anything that moves.
Can it handle document extraction?
Yes. Qwen3-VL models can analyze screenshots, scanned documents, receipts, and PDFs. They understand layout and content and return structured JSON — all through the same API endpoint.
Why not call the models directly?
One OpenAI-compatible endpoint, one API key, one bill. No glue code per model, no separate authentication. Makes it much easier to test, validate, and scale across the entire Qwen and Wan lineup.
Is this production-ready?
Over 150 enterprises run production workloads on our infrastructure. Backed by Alibaba Cloud, AWS, Google Cloud, and Azure. Cloudflare edge network for low-latency delivery worldwide.
Start building today.
$5 free credit — that's 166 images, 50 seconds of video, or thousands of API calls. No credit card required.