whatmodel.sh / video / HunyuanVideo 1.5
HunyuanVideo 1.5
↗ share
+ contribute
HunyuanVideo 1.5 is an open-weights video model from Tencent, released November 2025. Its published size is 8.3B parameters. Licence: tencent-hunyuan-community.
$ how it compares
The video models ranked either side of HunyuanVideo 1.5, out of 139 with a score, then more from Tencent.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 57 |  Kandinsky 5.0 Pro | 52±2 | api $0.040/sec +local | ||
| 58 |  HY-OmniWeaving | 52±4 | comparison only | ||
| 59 | HunyuanVideo 1.5 | 52±2 | api $0.075/sec +local | ||
| 60 |  Cosmos3-Nano | 52±3 | local | ||
| 61 |  PixVerse 5.5 | 52±10 | api $0.060/sec | ||
| more from Tencent | |||||
| 55 | HunyuanVideo 1.5 (I2V) | 53±2 | api $0.075/sec +local | ||
| 90 | HunyuanVideo-I2V | 45±4 | local | ||
| 105 | HunyuanVideo | 39±3 | api price unknown +local | ||
$ what it's good at
text to video▆▄▃▄▆▅█▃▂▂▂▄ #30 of 68 · 65% of its score · 5 tests
Beats Kling 2.1 Master here, though they rank higher overall. Trails Cosmos3-Nano here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| LMArena video | 1168.8 ✎ | ▂▂▃▆█▆·▆▅▁▂▇ beats 10/41 | 45% |
| Cosmos HUE human evaluation, text-to-video (Cosmos 3 report) | 86.5% ✎ | █▄▄··▄▄▄▄▄·▄ beats 4/9 | 8.5% |
| WorldJen human preference (Bradley-Terry) | 1462.0 ✎ | █···██·█··██ beats 1/5 | 6.5% |
| Sci-VBench testmini (expert human rating) | 46.8% ✎ | ▅·▃█▃·▅▃▅▃▃▃ beats 6/14 | 3.5% |
| HumanScore (T2V) | 91.1% ✎ | ▂··▂··█▄▂··▆ beats 10/11 | 1.5% |
image to video▁▂·▂▅█▅▄▅▆▃▄ #42 of 50 * · 30% of its score · 1 test
Trails Hailuo 01 (MiniMax video-01 / I2V-01), LongCat-Video, Cosmos3-Nano here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Arena image-to-video | 1197.8 ✎ | ▁▁▁▁▄▇▃▆▃▂▄█ beats 7/40 | 30% |
other tests1 not in a task · 4.2% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| VideoArgus-Bench T2V (rubric-grounded score) | 85.4% ✎ | █▄·▄█▄▄····▄ beats 6/8 | 4.2% |
$ where to run it
Through an API
api · fal.ai · $0.075/sec · fal.ai ↗ · price sourceUSD per second of generated video, 480p (the only resolution offered on this endpoint). Product: fal HunyuanVideo 1.5 text-to-video endpoint. API model: fal-ai/hunyuan-video-v1.5/text-to-video. Price basis: the page's endpoint billing record states billing_unit=seconds, price=0.075, and the sibling image-to-video page states "$0.075 per second"; the page's prose says "0.075 cents/s", which is taken as a typo for $0.075/s on the strength of the billing record. Currency: USD. fal.ai model page, observed 2026-09-25.
Other ways to run it
version
- Run it as published · needs ~20.42 GB, NVIDIA only · from github.com/Tencent-Hunyuan
torchrun --nproc_per_node=1 generate.py --prompt "A cat walks on the grass, realistic style." --resolution 480p --rewrite false --model_path ./ckpts
Official weights and documentation: huggingface.co/tencent ↗
$ how the score was worked out · LMArena leaderboard dataset 75%, Cosmos 3: Omnimodal World Mod… 9%, WorldJen: An End-to-End Multi… 6% 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 | 2 | 75% |
 Cosmos 3: Omnimodal World Models for Physical AI (arXiv:2606.02800v4) | 1 | 9% |
| WorldJen: An End-to-End Multi-Dimensional Benchmark for Generative Video Models | 1 | 6% |
| VideoArgus: Agentic Rubric-Grounded Unified Evaluation for Video Generation and Editing (arXiv:2608.05485v1) | 1 | 4% |
| Sci-VBench: Evaluating Knowledge- and Reasoning-Intensive Video Generation in Science Domains | 1 | 4% |
| HumanScore: Benchmarking Human Motions in Generated Videos | 1 | 1% |
| prior · pull toward the middle | 1% | |
| weighted average | = 51.9 |
$ 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