Kling 3.0 Standard (BudgetPixel)
The 85-credit baseline that still feels like Kling
- end_image
- image
- video
Kling 3.0 Standard is the budget tier of the Kling 3.0 video family, and "budget" is doing more work than you might expect. At 85 credits per second it's the cheapest Kling 3.0 on this pack's shelf - under Standard, only the older v2.6 Pro-style deals and the Turbo tier compete on price, and Standard undercuts Turbo's 120 while still giving you the full Kling 3.0 feature set: text-to-video, image-to-video with a start frame and an optional end frame, 3–15 second durations, and optional audio. The catch is resolution: Standard is the 720p tier.
That's the honest trade, and it's a good one. For the 90% of video work that lives on a phone screen or a website hero, 720p with end-frame control at 85 credits/second is a better deal than paying 4x for 4K pixels nobody will see. Standard is the node you use when the idea has to move now and you want the full Kling toolkit, not the Turbo speed special.
How it works
Standard BudgetPixel video flow: your prompt, optional start/end frames, and the audio toggle go into the Kling 3.0 Standard pipeline, the node polls until succeeded, and the finished MP4 downloads as a core VIDEO. Note that audio raises the per-second rate - the tooltip says so explicitly, and it's the same wrinkle as Pro, just at a lower base. The optional end frame is the feature that makes this tier genuinely useful rather than merely cheap.
The inputs that matter
- prompt - required, multiline.
- image - optional start frame.
- end_image - optional end frame. Free to use, and at this price it's the difference between "Kling wandered somewhere" and "Kling nailed the composition."
- generate_audio - optional; raises the per-second rate.
- length_seconds - 3–15. At 85 credits/sec, a 15-second clip is still a four-digit number, so run Cost Estimate with
length_secondsin the params. - aspect_ratio - framing.
Output is a single VIDEO.
Install
Manager (search "BudgetPixel") or:
cd ComfyUI/custom_nodes
git clone https://github.com/budgetpixel-ai/comfyui-budgetpixel.git
pip install -r comfyui-budgetpixel/requirements.txt
Restart, set BUDGETPIXEL_API_KEY or budgetpixel.json ({"api_key": "bpx_live_..."}) in the pack folder. No model files; requests is the only dependency.
Where people get burned
The trap is using Standard for a deliverable that needed 1080p or 4K. Standard is 720p, period - if your client asks for "4K, obviously," the cheap tier isn't a shortcut, it's a re-render bill on top of the wasted one. Know your resolution before you queue. Second, the audio surcharge: optional audio at Standard costs extra per second, so if you're going to strip the sound anyway, leave generate_audio off. Third, per-second billing means duration is the real price driver - a 15-second Standard run costs nearly as much as a 3-second 4K run, which is the weird math of this whole shelf. Use the Cost Estimate node and let the numbers pick your tier.
Inputs (7)
| Name | Type | Default | Description |
|---|---|---|---|
| prompt | STRING | Text description of the video. | |
| aspect_ratioopt | COMBO | 16:9 | Aspect ratio. Ignored when a start image already sets the frame. |
| end_imageopt | IMAGE | Optional end frame, used together with `image` (the start frame) to interpolate the video between the two frames. Same input forms as `image`. | |
| generate_audioopt | BOOLEAN | false | Generate audio with the video. Raises the per-second rate on Standard/Pro (no price impact at 4K) — estimate via POST /v1/cost. |
| imageopt | IMAGE | Optional start frame for image-to-video. Provide a public image URL, a data URI, raw base64, or an uploaded-file URL from POST /v1/uploads. Omit for text-to-video. | |
| length_secondsopt | INT | 53–15 | Video length in seconds (3–15). |
| force_rerunopt | BOOLEAN | false | — |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| video | VIDEO | — |