whatmodel.sh / video / Kling 3.0 Pro
Kling 3.0 Pro
↗ share
+ contribute
Kling 3.0 Pro is a closed-weights video model from Kuaishou, released February 2026.
Model information sources
$ how it compares
The video models ranked either side of Kling 3.0 Pro, out of 139 with a score, then more from Kuaishou.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 16 |  Wan 2.2 I2V Plus | 71±5 | api $0.10/sec | ||
| 17 |  Grok Imagine Video | 69±2 | api $0.050/sec | ||
| 18 | Kling 3.0 Pro | 69±2 | api $0.11/sec | ||
| 19 |  Veo 3.1 | 69±1 | api $0.40/sec | ||
| 20 |  Sora 2 Pro | 68±2 | comparison only | ||
| more from Kuaishou | |||||
| 40 | Kling 2.5 Turbo Pro | 57±2 | api $0.070/sec | ||
$ what it's good at
image to video▁▂·▂▅█▅▄▅▆▃▄ #14 of 50 · 70% of its score · 3 tests
Beats HappyHorse 1.1 here, though they rank higher overall. Trails Kling 3.0 here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Arena image-to-video | 1354.0 ✎ | ▁▁▁▁▄▇▃▆▃▂▄█ beats 25/40 | 44% |
| Rapidata physics realism (I2V, with reference) | 1123.0 ✎ | ▂··▂▂▆█▆▂▂▆▄ beats 11/19 | 25% |
| Rapidata camera movement (I2V) | 1062.2 ✎ | ▄···▄▄▄█··▄█ beats 5/8 | 1.2% |
other tests3 not in a task · 28% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Rapidata relative camera movement (I2V, alignment Elo) | 1000.0 ✎ | █···▄▄██·█▄█ beats 3/12 | 16% |
| Rapidata image-to-video realism (overall Elo) | 1290.9 ✎ | ▂······▃▂▃▃█ beats 10/12 | 8.4% |
| Datapoint Video Bench (overall Elo) | 1010.4 ✎ | ▃▃·▅·▃·▃··▃█ beats 5/9 | 3.8% |
$ where to run it
Through an API
api · fal.ai · $0.11/sec · fal.ai ↗ · price sourceUSD per second of generated video on fal.ai, audio off. Product: fal Kling 3.0 Pro text-to-video endpoint. API model: fal-ai/kling-video/v3/pro/text-to-video. Settings: generate_audio=false (endpoint default true costs $0.168/s; with voice control $0.196/s). Price basis: per-second rate printed by fal. Kling's own developer page lists the same $0.112/s for Kling 3.0 at 1080P without audio, but does not label a 'Pro' mode, so the explicitly named fal endpoint is used. Currency: USD. fal.ai model page, observed 2026-09-25.
$ how the score was worked out · LMArena leaderboard dataset 44%, Rapidata/world-model-physics… 25%, Rapidata/relative-camera-move… 16% and 3 more
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works
| source | results | share of the score |
|---|---|---|
 LMArena leaderboard dataset | 1 | 44% |
 Rapidata/world-model-physics published dataset-card rankings | 1 | 25% |
| Rapidata/relative-camera-movement-human-preference-900k published dataset-card rankings | 1 | 16% |
| Rapidata/image-to-video-realism published dataset-card rankings | 1 | 8% |
| Datapoint Video Bench leaderboard (datapointai/text-to-video-human-preferences-326k dataset card) | 1 | 4% |
| Rapidata/camera-movement published dataset-card rankings | 1 | 1% |
| prior · pull toward the middle | 2% | |
| weighted average | = 68.9 |
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json