whatmodel.sh / video / Sora (original, 2024)

Sora (original, 2024)

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

Sora (original, 2024) is a closed-weights video model from OpenAI, released December 2024.

Model information sources

Release date ↗

41.7±1.8comparison only: no current run option verifiedfrom 6 benchmark results across 5 sources

$ how it compares

The video models ranked either side of Sora (original, 2024), out of 139 with a score, then more from OpenAI.

#modelscorewhereRAMreleased
96Pyramid Flow (SD3) Peking University / Kuaishou42±3local~8GB2024-10
97Luma Dream Machine 1.0 Luma AI42±7comparison only2024-06
98Sora (original, 2024) OpenAI42±2comparison only2024-12
99Wan 2.1 T2V 1.3B Alibaba40±4api $0.040/sec +local~8.6GB2025-02
100Mochi 1 Genmo40±2api price unknown +local~6.5GB2024-10
more from OpenAI
20Sora 2 Pro OpenAI68±2comparison only2025-10
27Sora 2 OpenAI65±2comparison only2025-10

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

$ what it's good at

text to video▇▅▄▆█▆▇▂▃▃▂▄ #56 of 94 · 86% of its score · 4 tests

Beats Open-Sora 2.0, Pyramid Flow (SD3) here, though they rank higher overall.

testresultvs the others in its rundifficultycountswhy
LMArena video1068.9 ✎▂▃▃▆█▆·▆▅▁▂▇ beats 3/41avg 1147▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 53%typical
AIGVE-60K human perception MOS (LOVE)62.1% ✎▄▄·█··█▄▄▄▄▄ beats 10/10easy · avg 73%▇▇▇▇▇▇▇ 17%at its level
AIGVE-60K human text-video correspondence MOS (LOVE)59.7% ✎▄▄·█▄█▄··▄▄▄ beats 10/10easy · avg 70%▇▇▇▇ 10%at its level
T2VPhysBench (human physical consistency)28.0% ✎█··█·█·····█ beats 0/3medium · avg 46%▇▇▇ 6.3%at its level
image to video▁▁▃▂██▄▃▃▄▂▂ #69 of 74 * · 6.7% of its score · 1 test
testresultvs the others in its rundifficultycountswhy
MagicArena image-to-video Elo (Seaweed-7B report)903.0 ✎▄·····▄█··▄▄ beats 0/5avg 1043▇▇▇ 6.7%typical
other tests1 not in a task · 6.4% of its score
testresultvs the others in its rundifficultycountswhy
Wan-Bench (Wan2.1 release) weighted score70.0% ✎█·······█··█ beats 1/2easy · avg 78%▇▇▇ 6.4%at its level · small run

$ where to run it

No verified API or download recorded yet · add one

$ how the score was worked out · LMArena leaderboard dataset 53%, LOVE: Benchmarking and Evalua… 26%, Seaweed-7B: Cost-Effective Tr… 7% and 2 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

sourceresultsshare of the scorelicence
LMArena leaderboard dataset1▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 53%CC BY 4.0
LOVE: Benchmarking and Evaluating Text-to-Video Generation and Video-to-Text Interpretation (AIGVE-60K)2▇▇▇▇▇▇▇▇ 26%CC-BY-4.0
Seaweed-7B: Cost-Effective Training of Video Generation Foundation Model1▇▇ 7%CC-BY-4.0
Wan-AI/Wan2.1-T2V-14B model card1▇▇ 6%Apache-2.0
T2VPhysBench: A First-Principles Benchmark for Physical Consistency in Text-to-Video Generation1▇▇ 6%CC-BY-4.0
prior · pull toward the middle▇ 1%
weighted average= 41.7

$ 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