ComparisonsKling vs Happy Horse: how do Kling 3.0 and Happy Horse 1.1 differ?
Makernice Team · ·
Prices as of November 12, 2026 (price list v15).
Key takeaways
- Happy Horse 1.1 and every Kling 3.0 variant accept 3 to 15 seconds per clip and prompts of up to 2,500 characters.
- Happy Horse 1.1 has reference-to-video with up to 9 images but no first and last frame; Kling 3.0 Pro and Standard have first and last frame but no reference-to-video.
- Happy Horse 1.1 picks 720p or 1080p with the resolution field; Kling 3.0 fixes the tier by model id, from 720p on Standard to 4k on the 4K variant.
- Happy Horse 1.1 accepts 4:3 and 3:4 in text-to-video; Kling 3.0 accepts 9:16, 16:9 and 1:1 only.
Happy Horse 1.1 from Alibaba and Kling 3.0 from Kuaishou have the same clip length range, 3 to 15 seconds, and the same 2,500 character prompt limit on Makernice. What differs is the set of modes: Happy Horse 1.1 offers reference-to-video and Kling 3.0 offers first and last frame, and neither offers both. They also choose resolution differently and accept different aspect ratios. This page compares catalog facts and prices only; it does not say which output looks better.
What do Happy Horse 1.1 and Kling 3.0 share?
Happy Horse 1.1 and Kling 3.0 share a 3 to 15 second duration with a default of 5, a 2,500 character prompt cap, text-to-video and image-to-video, a 720p tier and a 1080p tier, and the 9:16, 16:9 and 1:1 ratios in text-to-video. A plain text-to-video request inside those bounds, or an image-to-video request with a JPEG or PNG start image, runs on both after a change of model id and, for 1080p, of the resolution field.
Field-by-field specs
Happy Horse 1.1 is one model id with a resolution field and reference-to-video, while Kling 3.0 is three model ids with fixed tiers and first and last frame. The table lists every catalog field for the four ids.
| Spec | Happy Horse 1.1 | Kling 3.0 Standard | Kling 3.0 Pro | Kling 3.0 4K |
|---|---|---|---|---|
| Vendor | Alibaba | Kuaishou | Kuaishou | Kuaishou |
| Modes | t2v, i2v, r2v | t2v, i2v, flf | t2v, i2v, flf | t2v |
| Duration | 3 to 15 s, default 5 | 3 to 15 s, default 5 | 3 to 15 s, default 5 | 3 to 15 s, default 5 |
| resolution field | 720p, 1080p | Not accepted | Not accepted | Not accepted |
| Default tier | 720p by default | Fixed at 720p | Fixed at 1080p | Fixed at 4k |
| Text-to-video ratios | 16:9, 9:16, 1:1, 4:3, 3:4 | 9:16, 16:9, 1:1 | 9:16, 16:9, 1:1 | 9:16, 16:9, 1:1 |
| First and last frame | No | Yes, no aspect_ratio field | Yes, no aspect_ratio field | No |
| Reference images | Up to 9 (r2v) | No r2v | No r2v | No r2v |
| Prompt limit | 2,500 characters | 2,500 characters | 2,500 characters | 2,500 characters |
| Audio | Not listed | Always on, no switch | Always on, no switch | Always on, no switch |
Model pages: Happy Horse 1.1, Kling 3.0 Standard, Kling 3.0 Pro and Kling 3.0 4K.
Does Happy Horse 1.1 or Kling 3.0 have first and last frame?
Kling 3.0 Pro and Kling 3.0 Standard do, and Happy Horse 1.1 does not. The two Kling 3.0 variants take a start image as image_url and an end image as last_image_url and generate the motion between them. Happy Horse 1.1 rejects mode flf, so a shot defined by its first and last frames needs Kling 3.0 or another model with that mode, such as Wan 3.0.
Which reads reference images, Happy Horse 1.1 or Kling 3.0?
Happy Horse 1.1, with up to 9 images in reference-to-video. Each reference image needs a short side of at least 400 pixels, and the mode takes no aspect_ratio field. Kling 3.0 has no reference-to-video on any variant; Kling 4.0 adds it with up to 10 images.
How do you get 1080p on Happy Horse 1.1 and Kling 3.0?
On Happy Horse 1.1, send resolution set to 1080p; without it the clip renders at 720p. On Kling 3.0, call kling-3.0-pro, which is fixed at 1080p and rejects a resolution field. For a higher tier, Kling 3.0 4K renders at 4k, and only for text-to-video.
What does the catalog say about Happy Horse audio?
The Makernice catalog does not list audio behavior for Happy Horse 1.1, so this page makes no claim either way; check a short test clip before building on it. Kling 3.0 is documented as returning audio on every request, with no switch.
How much do Happy Horse 1.1 and Kling 3.0 cost per second?
On Makernice, Happy Horse 1.1 is $0.168 per second at 720p and Kling 3.0 Standard is $0.1209 per second, on price list v15 (November 12, 2026). Each tier has one per-second rate across its modes. The 3 s column is the shortest clip both lines accept. Clip price = per-second price × seconds, rounded to the cent.
| Model · tier | Per second | 3 s | 10 s |
|---|---|---|---|
| Happy Horse 1.1 · 720p | $0.168 | $0.50 | $1.68 |
| Kling 3.0 Standard · 720p | $0.1209 | $0.36 | $1.21 |
| Happy Horse 1.1 · 1080p | $0.216 | $0.65 | $2.16 |
| Kling 3.0 Pro · 1080p | $0.1612 | $0.48 | $1.61 |
| Kling 3.0 4K · 4k | $0.5142 | $1.54 | $5.14 |
For Happy Horse 1.1 against a ByteDance model, see Happy Horse 1.1 vs Seedance 2.5 and Seedance 2.0 vs Happy Horse 1.1.
Which should I choose, Kling or Happy Horse?
Pick by mode, because no side-by-side quality test of the two is published here: Happy Horse 1.1 for reference images or 4:3 and 3:4 text-to-video. Kling 3.0 fits first and last frame and documented audio on every clip, and Kling 3.0 4K adds text-to-video at 4k. Requests that fit both lists run on either one.
Frequently asked questions
Does Happy Horse 1.1 support reference images?
Can Kling 3.0 generate 4:3 video?
Is there a Happy Horse 4K option?
Do both models accept the same prompt length?
Models in this post
VideoHappy Horse 1.1
Alibaba's Happy Horse 1.1: a focused model for text, image and reference-driven clips — 3–15 seconds at 720p or 1080p, up to 9 references.
VideoKling 3.0 Standard
-4%Kling 3.0 with audio at the 720p tier — the budget pick for social clips that need sound.
VideoKling 3.0 Pro
-4%Video that ships with sound: Kuaishou's Kling 3.0 Pro adds audio to every 3–15 second clip, from text, a still or first and last frames. 1080p tier.
VideoKling 3.0 4K
Prompt-to-4K with sound: Kling 3.0 4K turns text into 3–15 second clips with audio, priced at the 4K tier.