whatmodel.sh / video / P-Video

P-Video

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

P-Video is a closed-weights video model from Pruna, released February 2026.

Model information sources

Release date ↗

55.0±1.8#38 of 93 video modelsfrom 3 benchmark results across 2 sources

$ how it compares

The video models ranked either side of P-Video, out of 139 with a score.

#modelscorewhereRAMreleased
44Kling O1 Pro Kuaishou (Kling AI)56±2api $0.11/sec
45Wan 2.6 Flash Alibaba55±7api $0.025/sec2026-01
46P-Video Pruna55±2api $0.020/sec2026-02
47Vidu Q2 Turbo ShengShu (Vidu)55±3api $0.050/sec2025-09
48Hailuo 2.3 MiniMax55±2api $0.047/sec2025-10

see all video models · find the ones that fit your machine

$ what it's good at

text to video▆▄▃▄▆▅█▃▂▂▂▄ #23 of 68 * · 58% of its score · 1 test

Beats Seedance 1.0 Pro, Kling O1 Pro here, though they rank higher overall.

testresultvs the others in its rundifficultycountswhy
LMArena video1207.3 ✎▂▂▃▆█▆·▆▅▁▂▇ beats 17/41avg 1147▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 58%typical
image to video▁▂·▂▅█▅▄▅▆▃▄ #34 of 50 * · 41% of its score · 2 tests

Beats Seedance 1.5 Pro here, though they rank higher overall. Trails Cosmos3-Nano, Hailuo 2.3, Vidu Q2 Turbo here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
Arena image-to-video1243.6 ✎▁▁▁▁▄▇▃▆▃▂▄█ beats 14/40avg 1189▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 38%typical
Physics-IQ Verified (I2V, best-practice prompts)25.3% ✎██···█··█·██ beats 0/5hard · avg 25%▇ 2.4%blunt · at its level

$ where to run it

Through an API

api · Replicate · $0.020/sec · replicate.com ↗ · price sourceUSD per second of output video, 720p standard (draft mode off). Product: Replicate model prunaai/p-video (Pruna's official model on Replicate). Price basis: Replicate per-output-second tariff (not hardware time). 720p draft $0.005/s; 1080p $0.04/s standard, $0.01/s draft. The page notes paid pricing resumes after an extended free launch period; the listed paid rate is recorded. Currency: USD. Replicate model page, observed 2026-09-25.

$ how the score was worked out · LMArena leaderboard dataset 96%, Physics-IQ Verified 2%

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

sourceresultsshare of the scorelicence
LMArena leaderboard dataset2▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 96%CC BY 4.0
Physics-IQ Verified1▇ 2%CC-BY-4.0
prior · pull toward the middle▇ 1%
weighted average= 55.0

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

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