What GPU do I need to run zai-org/CogVideoX-5b?
5.6B parameters, published in BF16. View on Hugging Face
Set up CogVideoX-5bCogVideoX-5b is published by zai-org on Hugging Face, with 16,229 downloads and 686 likes to date. It's a unlisted-architecture model built for text-to-video, published natively in BF16.
VRAM required & cheapest live GPU fit
Required VRAM = weight size at each precision, plus a fixed overhead for activation memory and allocator fragmentation. Diffusion and video models carry no KV-cache. The real driver of extra memory is output resolution and frame count, which this flat overhead does not model. Full formula and assumptions: methodology.
| Precision | Weight size | Required VRAM | Cheapest live fit | GPUs needed | Est. $/hr (full fit) |
|---|---|---|---|---|---|
| BF16 | 10.4 GB | 12.5 GB | RTX A4000 | 1 | $0.167/hr |
| FP8 (quantized) | 5.2 GB | 6.2 GB | RTX 4000 SFF Ada | 1 | $0.198/hr |
| INT4 (quantized) | 2.6 GB | 3.1 GB | RTX 3060 | 1 | $0.110/hr |
A GPU is only matched to a row if its hardware supports that precision, and the primary recommendation is always a single-GPU fit when one exists.
INT4 requires a quantized checkpoint actually published for this model, check its Hugging Face page before relying on this row.
Cheapest way to run CogVideoX-5b at its published (BF16) precision: 1× RTX A4000, at $0.167/hr per GPU ($0.167/hr total). Quantizing to FP8 or INT4 (rows above) can cost less, but requires a compatible quantized checkpoint to exist for this model.
CogVideoX-5b: common questions
Does CogVideoX-5b fit on a 16 GB GPU?
Yes. At BF16 it needs 12.5 GB of VRAM, so a 16 GB card holds it with 3.5 GB to spare. A 12 GB card is not enough for it at BF16.
What is the least VRAM CogVideoX-5b can run in?
3.1 GB, at INT4 (quantized), which fits a 6 GB card, against 12.5 GB at BF16. That row assumes an INT4 checkpoint has actually been published for this model, so check its Hugging Face page before planning around it.
Does quantizing CogVideoX-5b lower the GPU bill?
Yes. At BF16 the cheapest live fit is one RTX A4000 at $0.167/hr. At INT4 (quantized) it drops to one RTX 3060 at $0.110/hr, provided a quantized checkpoint exists for it.
Weight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.
More CogVideoX models
- CogVideoX-5b-I2V (5.6B, BF16)
- CogVideoX-2b (1.7B, F16)
Alternatives at this size
Other models for text-to-video within about a third of CogVideoX-5b's 5.6B parameters, from other model lines.
- Wan2.2-TI2V-5B-Diffusers (5.0B, F32)
- FastWan2.2-TI2V-5B-FullAttn-Diffusers (5.0B, BF16)
More on CogVideoX-5b
Fits on a 16 GB GPU at BF16: every model that fits in 16 GB.
Fits on an 8 GB GPU at FP8: every model that fits in 8 GB.
Best video generation models: how CogVideoX-5b ranks against the rest.
Related reading: RTX A4000 pricing and specs, Fast diffusion inference on GPU VMs, and What AI inference is.
