Loading model…正在加载模型详情…
KlingGenerate video from text with Kling 2.5 Turbo; Kuaishou Kling emphasizes prompt following, fluid motion and consistent visual style.
- From $0.26 / clipSave up to 25%
- 5s or 10s
Examples
Real-time playback. Wide shot of a ruined city: collapsed towers, fires blazing, storm clouds with lightning. Camera drops fast from the sky over burning streets and tilted buildings. Smoke and dust fill the air. A lone hero walks out of the ruins, silhouetted by fire. Camera shifts front: his face is dirty with dust and sweat, eyes firm, a faint smile. Wind blows, debris rises. Extreme close-up: his eyes reflect the approaching enemy. Music and drums hit. Final wide shot: fire forms a blazing halo behind him — reborn in flames with epic cinematic vibe.
Aerial IMAX Wide: Sunlit alpine mountains sparkle with untouched snow. POV Descent: Skis carve into powder, snow blasts outward, speed overwhelming. Tracking Chase: Drone follows tightly at high velocity, motion ultra-stable. Slow Motion: Launch off natural snow ramp, snowflakes scatter like diamonds in sunlight. Epic Finale: Crane shot pulls wide, skier carving a perfect line across vast slopes, breathtaking cinematic scale.
Real-time speed. Inside a swaying tram, camera extremely close to a stunning teenage girl sitting by the window. Pale skin, long blonde hair over her shoulders, wired earphones, eyes closed, calm expression, lips slightly pressed. Sunlight streams through the window, casting a soft patch on her eyelids. Other passengers blur past, faint sounds of metal tracks and bell in the background. Her hair sways gently with the tram, then she slowly opens her eyes, lashes trembling, gaze focusing on the outside, thoughtful. Realistic, natural light, warm tones, shallow depth of field blurs background, highlighting her face and earphone. Camera fixed in front, slowly pushing from close-up to extreme close-up, focus on lashes and eyes, capturing subtle changes as she opens her eyes, full of immersive emotion and tension.
What is Kling 2.5 Turbo?
Kling 2.5 Turbo is a video generation model from Kuaishou Kling. Kuaishou Kling’s description highlights understanding of scene instructions, smooth movement and consistent artistic styles.
UniRoute offers the text-to-video variant with a choice of clip duration. Describe the scene you want and select a duration; the request is quoted as a whole before you run it.
Kling 2.5 Turbo follows earlier Kling models such as Kling V2.1. Kuaishou Kling reports improvements in prompt adherence, motion and physical interactions; use the examples below to explore its approach to different scenes.
What you can do
What you can do with Kling 2.5 Turbo through UniRoute today.
Text to video
Describe a subject, sequence of actions and camera movement, then choose a duration to create a clip from the prompt.
Other Kling 2.5 Turbo modes (image-to-video) are not available on UniRoute yet; the current text-to-video variant has no reference image input.
Specs & variants
The variants you can call, with their options and prices.
| Variant | Model ID | Modes | Inputs | Output | Price |
|---|---|---|---|---|---|
| Kling 2.5 Turbo Text to Video | kling/v2-5-turbo-text-to-video-pro | Text to video | Text prompt and an option for duration | Duration: 5s, 10s | $0.26 to $0.53 per request |
Pricing
Priced per request. The option you choose sets the price.
| Option | Official API | UniRoute | Save |
|---|---|---|---|
| 5s | $0.35 | $0.2625 | Save 25% |
| 10s | $0.70 | $0.525 | Save 25% |
Official price: kling.ai
Prices updated 2026-11-08.
Pay as you go from a single USD wallet balance. You see the quote before each request.
API
Call Kling 2.5 Turbo from your own code with the media API.
cURL
# Requires bash, curl and jq. Set API_KEY to your own unified UniRoute Key.set -euo pipefail: "${API_KEY:?Set API_KEY first}"BASE='https://api.uniroute.app/v1/media'MODEL='kling/v2-5-turbo-text-to-video-pro'OPTION='5'PROMPT='A calm sunrise over a mountain lake'curl -fsS --max-time 60 "$BASE/catalog" -H "Authorization: Bearer $API_KEY" | jq -e '.data.models'BODY=$(jq -n --arg model "$MODEL" --arg option "$OPTION" --arg prompt "$PROMPT" '{model:$model,option:$option,prompt:$prompt}')QUOTE_BODY=$(printf '%s' "$BODY" | jq '{model,option}')QUOTE=$(curl -fsS --max-time 60 "$BASE/quote" -H "Authorization: Bearer $API_KEY" -H 'Content-Type: application/json' -d "$QUOTE_BODY")CAP=$(printf '%s' "$QUOTE" | jq -er '.data.quoted_usd')BODY=$(printf '%s' "$BODY" | jq --arg cap "$CAP" '. + {max_cost_usd:$cap}')IDEMPOTENCY="example-$(date +%s)-$$"create() { curl -fsS --max-time 60 "$BASE/tasks" -H "Authorization: Bearer $API_KEY" -H "Idempotency-Key: $IDEMPOTENCY" -H 'Content-Type: application/json' -d "$BODY"; }TASK_ID=$(create | jq -er '.data.task_id')# The same request and idempotency key must return the original task without charging again.REPLAY_ID=$(create | jq -er '.data.task_id')[ "$TASK_ID" = "$REPLAY_ID" ]for attempt in $(seq 1 120); do TASK=$(curl -fsS --max-time 60 "$BASE/tasks/$TASK_ID" -H "Authorization: Bearer $API_KEY") printf '%s' "$TASK" | jq -e '.data | {task_id,status,billing}' if printf '%s' "$TASK" | jq -e '.data.status == "success" and .data.billing.state == "charged"' >/dev/null; then index=0 while IFS= read -r url; do index=$((index+1)) curl -fL --max-time 120 "$url" -o "$TASK_ID-$index.bin" done < <(printf '%s' "$TASK" | jq -er '.data.urls[].url') break fi if printf '%s' "$TASK" | jq -e '.data.status == "failed" and .data.billing.state == "released"' >/dev/null; then exit 1; fi if [ "$attempt" = 120 ]; then echo "Polling limit reached. Keep task ID $TASK_ID and query it later; do not create a new task." >&2 exit 1 fi sleep 5donecurl -fsS --max-time 60 "$BASE/tasks?limit=20" -H "Authorization: Bearer $API_KEY" | jq -e '.data'# Check this task and the caller Key in Console usage; billing.charged_usd is the actual USD charge.Parameters
| Name | In | Description |
|---|---|---|
model | Body | Model ID to run. |
option | Body | One of the options offered for the model: 5, 10. |
prompt | Body | Text prompt describing what to generate. |
max_cost_usd | Body | Maximum cost in USD for this task, taken from the quote. |
Idempotency-Key | Header | A unique key per task, so a repeated request returns the original task instead of charging again. |
Guides
Use cases
Typical ways to put Kling 2.5 Turbo to work.
Action and camera studies
Explore an action scene or camera move from a written shot description before planning the production.
Mood and character shots
Try a close-up, a quiet interaction or a stylized setting to find a visual direction for a scene.
Social and advertising drafts
Generate video concepts from a short creative brief, then refine the subject, movement and atmosphere in the prompt.
Prompt-based video products
Add video creation to an application with the media API, using a quote and an asynchronous task for each request.
FAQ
Common questions about Kling 2.5 Turbo.
What is Kling 2.5 Turbo?
Kling 2.5 Turbo is Kuaishou Kling’s AI video model. On UniRoute you can use its text-to-video variant from this page or through the media API.
Which settings can I choose?
The callable variant accepts a text prompt and a duration option. The specs and pricing tables show the available choices without implying a selectable resolution.
How much does Kling 2.5 Turbo cost?
From $0.26 / clip. The price covers the whole clip at the duration you choose. The full table is in the Pricing section, with prices updated 2026-11-08.
Can I reuse one of the example prompts?
Yes. Choose “Use this” on an example to copy its prompt into the workspace, then adjust the wording and duration before generating.
How do I call it from code?
Use the media API to quote the model and option, create a task with an Idempotency-Key and a maximum cost, and poll until the video is ready. The API section shows the request flow.
Is there a free trial or free credits?
No. Generation is paid from your wallet balance. You can open the workspace and review the price without paying, create an account to see a live quote for your exact settings, and you are charged only for requests that are accepted.
Related models
Other versions, models from the same provider and alternatives.
