kling-video-o3/pro/reference-to-video
Pro-tier reference-driven 3-15s video generation from $0.112 per second.
Upload a start still and an end still, add a written brief for what happens in between, and get a short clip with optional synchronized native audio. On RunComfy you can pin both keyframes in the model UI, then reuse the same Input over HTTP API.
Output format: Video / start + end images / 5–20 seconds / 720p or 1080p / optional native audio
| Parameter | Required | Type | Default | Range / Options | Description |
|---|---|---|---|---|---|
| start_image_url* | Yes (*) | string | — | PNG, JPEG, WebP URL | Image used as the first frame of the video. |
| end_image_url* | Yes (*) | string | — | PNG, JPEG, WebP URL | Image used as the last frame of the video. |
| prompt* | Yes (*) | string | — | — | Text describing the transition and motion between the two frames. |
| aspect_ratio | No | string | auto | auto, 21:9, 2:1, 16:9, 4:3, 1:1, 3:4, 9:16 | Output frame shape; auto lets the model choose. |
| resolution | No | string | 720p | 720p, 1080p | Output resolution. |
| duration | No | integer | 5 | 5–20 | Clip length in seconds; required in practice so the end frame can be placed. |
| generate_audio | No | boolean | true | true, false | Whether to generate synchronized audio. |
| safety_tolerance | No | integer | 2 | 0–4 | 0 is strictest; 4 is most permissive. |
Billing for FLUX 3 First Last Frame to Video is based on the length of the generated clip and the selected resolution.
| Resolution | Rate |
|---|---|
| 720p | $0.17 per second |
| 1080p | $0.29 per second |
Estimated cost examples (720p)
| Duration | Approx. cost |
|---|---|
| 5 s | ~$0.85 |
| 10 s | ~$1.70 |
| 20 s | ~$3.40 |
Estimated cost examples (1080p)
| Duration | Approx. cost |
|---|---|
| 5 s | ~$1.45 |
| 10 s | ~$2.90 |
| 20 s | ~$5.80 |
Pro-tier reference-driven 3-15s video generation from $0.112 per second.
Easily add custom LoRA for unique styles and effects.
Turn reference images into smooth 720P or 1080P video with one prompt.
Turn static images into fluid, realistic 1080p motion with smart style control.
Create lifelike speech-synced visuals from scripts or clips with Kling Lipsync for precise facial animation and realistic results.
Generates up to 4-minute songs with vocals and lyrics from text tags
FLUX 3 First Last Frame to Video takes a start image, an end image, and a text prompt, then generates the motion between those two keyframes with optional synchronized native audio. It fits pose-to-pose transitions, product reveal arcs, and scene bridges when both bookend frames are already locked.
FLUX 3 First Last Frame to Video locks both the opening and closing stills, so you control the start and end of the clip. FLUX 3 Image to Video only requires a start image and continues motion from that single frame without a required end frame.
Yes. Generate Audio is on by default, so FLUX 3 First Last Frame to Video can output synchronized speech, SFX, and ambience with the picture. Turn it off when you need silent video.
Duration is a whole number of seconds from 5 to 20. An explicit length helps place the end frame correctly for your transition pacing.
Resolution options are 720p (default) and 1080p. Aspect ratio defaults to auto, or you can lock 21:9, 2:1, 16:9, 4:3, 1:1, 3:4, or 9:16 for known delivery formats.
start_image_url, end_image_url, and prompt are required. Optional fields are aspect_ratio, resolution, duration (5–20, default 5), generate_audio, and safety_tolerance (0–4, default 2). Start and end images should be PNG, JPEG, or WebP URLs.
Prototype in the RunComfy model UI, then call the same template through the RunComfy API using identical Input fields (start_image_url, end_image_url, prompt, aspect_ratio, resolution, duration, generate_audio, safety_tolerance). Validate prompts and keyframes in the UI first, then use your account API key and credits for automated jobs.
Billing is based on generated video duration and resolution: $0.17 per second at 720p, and $0.29 per second at 1080p. See the pricing shown on this page for the latest rates.
RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.
