VASA-1
No estimate
No hardware requirements for this model
The weights for this model have not been published, so it cannot be downloaded or run on your own hardware at any size. It is reachable only through its provider, and no graphics card changes that.
On record
Full specification
Everything on record for this model. Most of it describes how it was trained rather than how it runs — useful context for judging how much work went into it, and how it compares with models built at a different scale.
Origin
Who built this model, where, and when it was published.
- Organisation
- Microsoft Research Asia
- Organisation type
- Industry
- Country
- China
- Published
- 31 October 2024
- Authors
- Sicheng Xu, Guojun Chen, Yu-Xiao Guo, Jiaolong Yang, Chong Li, Zhenyu Zang, Yizhong Zhang, Xin Tong, Baining Guo
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Video, Audio
- Task
- Video generation
Size
How large the model is and how much data it was trained on. Parameters are the figure that decides whether it fits on a given graphics card.
- Parameters
- 229M
- Training data
- tokens
"The parameter counts of our 3D-aided face latent model and diffusion transformer model are about 200M and 29M respectively"
"The total data used for training comprises approximately 500K clips, each lasting between 2 to 10 seconds. "
Training compute
The arithmetic performed to train the model, measured in floating-point operations. It is a measure of what the training run cost, not of how fast the finished model answers you.
- Training compute
- 4 × 10¹⁹ FLOP
- How it was established
- Hardware
38700000000000 FLOP/GPU/sec * 240 hours * 3600 sec / hour * 4 GPUs * 0.3 [assumed utilization] = 40124160000000000000 FLOP
The training run
What it physically took to train: which chips, how many, for how long, and what that drew from the wall.
- Training hardware
- NVIDIA RTX A6000
- Chips used
- 4
- Wall-clock time
- 240 hours (10 days)
- Power draw
- 2.4 kW
"Our face latent model takes around 7 days of training on a 4 NVIDIA RTX A6000 GPUs workstation, and the diffusion transformer takes around 3 days." 10 days = 240 hours
Availability
Whether you can obtain the model and run it on your own hardware, which is what decides if any of the graphics-card figures on this page apply.
- Weights
- Closed — provider access only
- Model access
- Unreleased
- Training code
- Unreleased
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Record confidence
- Confident
Sources
Where this record came from and when it was last checked.
- Reference
- VASA-1: Lifelike Audio-Driven Talking Faces Generated in Real Time
- Last updated
- 28 November 2025
What the numbers mean
Where it came from
VASA-1 was published by Microsoft Research Asia, in the country recorded as China, during October 2024. It comes out of an organisation categorised as industry.
It works in the domain of Video, Audio, and is recorded as performing the task of video generation.
Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.
What went into building it
Training it took a computation budget of roughly 4 × 10¹⁹ FLOP, on hardware recorded as NVIDIA RTX A6000. That figure measures what producing the model cost, and has no bearing on how fast it answers.
Answers
VASA-1 — common questions
VASA-1— when was it released?
It was published in October 2024. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
VASA-1— what is it used for?
It works in the domain of Video, Audio, and is recorded as handling the task of video generation. These are the areas it was designed around; they describe intent rather than a hard boundary.
VASA-1— how much compute was used to train it?
Training consumed around 4 × 10¹⁹ FLOP, on hardware recorded as NVIDIA RTX A6000. That measures what producing the model cost and says nothing about how quickly it answers once trained — inference speed comes from memory bandwidth, not from the training budget.
VASA-1— what GPU do I need to run it?
None. This is a closed model — its weights were never published, so it cannot be downloaded or run on your own hardware at any price. It is reachable only through its provider.
VASA-1— is it open source?
No. Its weights have not been published, so it exists only as a service controlled by its owner.
VASA-1— how many parameters does it have?
It has a parameter count of 229M. "The parameter counts of our 3D-aided face latent model and diffusion transformer model are about 200M and 29M respectively". That figure is the total, and it is what decides how much memory the model needs — roughly half a gigabyte per billion at the compression most people use.
VASA-1— who created it?
It was published by Microsoft Research Asia, based in China, an organisation categorised as industry.
The other direction
Looking at it from the other side?
This page starts from the model. If you already own a card and want to know everything it will run, start from the hardware instead.