Introduction
Wan 2.1 Image LoRA Text to image renders one photoreal frame from Alibaba's open 14B video model with up to three of your Wan LoRA files loaded, in nine shapes at 1k or 1.5k, at one flat price and without a content filter.
Choose a mode by starting material
| What you have | Mode |
|---|---|
| a written description | Wan 2.1 Image LoRA Text to image |
| an image to change | Wan 2.1 Image LoRA Image editing |
| a scene that should move | Wan 2.1 LoRA Text to video |
Key features & highlights
Video LoRAs on stills
Wan 2.1 renders an image as a single frame, so Wan 2.1 14B text-to-video LoRAs apply here. In our example an origami LoRA turned a masked couple into folded-paper figures; without it, at the same seed, they stayed people.
1536 pixels at no extra cost
The 1.5k tier puts 1536 pixels on the long edge and costs the same as 1k. In our runs 1536 x 1536 rendered cleanly, without doubled people.
Wan 2.2 style LoRAs
A single-file Wan 2.2 style LoRA turned a crisp colour photograph into warm brown film light on this model.
Uncensored with custom weights
LoRAs are not reviewed and the output is not filtered.
Request parameters
| Parameter | Required | Description |
|---|---|---|
prompt | Yes | Trigger words first if your LoRA has them, then the subject, the setting and the light. Up to 5,000 characters. at least 1 characters · at most 5000 characters |
aspect_ratio | No | 1:1 (the default), 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 21:9 or 9:21. At 1k: 1024 x 1024, 1344 x 768, 1152 x 896, 1216 x 832, 1344 x 576 and their portrait swaps. |
resolution | No | 1k (about one megapixel, the default) or 1.5k (1536 pixels on the long edge, for example 1536 x 864 for a landscape frame). Same price for both. |
loras | No | Up to three {path, scale} entries trained for Wan 2.1 14B; path is a public direct link to a .safetensors file, scale runs from 0 to 4 (1 by default). Default: |
output_format | No | jpeg (the default), png or webp. No effect on price. |
seed | No | Use it to compare runs; the same seed did not always return the same image in our test. |
Pricing
$0.03 / image
One flat price per image, whatever the size, tier or LoRAs.
Use cases
LoRA previews
Check what a character or style LoRA does on a still before rendering clips: a still costs a fraction of a clip and comes back in seconds.
Opening frames
Make the first frame here with the same LoRA, then animate it with Wan 2.1 LoRA image-to-video so the look stays consistent from still to clip.
Platforms that need unfiltered stills
Load the LoRA your audience expects; everyone shown must be an adult.
Prompt guide & pro tips
Write like a photographer
Lens, light and colour words steer the look well: "50mm, shallow depth of field, warm lantern light" did more for our images than adjectives about quality.
Use 1.5k
Same price, more pixels, and no doubled people in our runs. Both sides are rounded down to a multiple of 32.
Trigger words first
Put a LoRA's trigger phrase at the start of the prompt. On stills some trigger words already push the style by themselves, so compare against a run without the LoRA.
Direct links only
A page link or a gated file fails with a clear message and is not charged.
API quick start
# Generate the idempotency key once; reuse it when retrying after a lost response.
export SPICY_IDEMPOTENCY_KEY="${SPICY_IDEMPOTENCY_KEY:-$(uuidgen)}"
curl -X POST https://api.spicyapi.ai/api/v1/jobs/createTask \
-H "Authorization: Bearer $SPICY_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: $SPICY_IDEMPOTENCY_KEY" \
-d '{
"model": "alibaba/wan-2.1-image-lora/text-to-image",
"input": {
"aspect_ratio": "1:1",
"output_format": "jpeg",
"prompt": "A ceramic teapot on a linen cloth, soft window light, muted palette",
"resolution": "1k"
}
}'
# Copy data.taskId from the response into TASK_ID, then poll the status.
TASK_ID="TASK_ID_FROM_RESPONSE"
curl "https://api.spicyapi.ai/api/v1/jobs/recordInfo?taskId=$TASK_ID" \
-H "Authorization: Bearer $SPICY_API_KEY"
See the API reference for authentication, the complete schema, polling, and error handling.
