Wan2.2-T2V-A14B
Official · BF16 · safetensors
Runtime certified: Wan Diffusers
| Precision / quantization | Format | Bits | Publisher trust | Deployment state |
|---|---|---|---|---|
| BF16 | safetensors | 16 | Official | Runtime certified |
Wan · General · Wan2.2
Wan2.2 T2V A14B: Specs, GPU requirements, hosting & deployment
Wan2.2 T2V A14B is a current Video Generation release in the Wan family, certified from its immutable official source revision.
Total parameters27.0B
Active parameters14.0B
Context windowN/A / Not directly comparable
Release statusCurrent · Current
ArchitectureVideo diffusion MoE transformer
ModalitiesText, Video
LicenseApache License 2.0
Deployment stateDeployable with certified configuration
Commercial stateCommercial Use Allowed
Release date2025-07-24
Architecture, strengths and use cases
Wan2.2 T2V A14B is an official self-hostable Alibaba Wan Video generation release. It uses Video diffusion MoE transformer with 27B / 14B MoE. Exact checkpoint, license, runtime, and hardware evidence is tied to the verified upstream revision.
Wan2.2 T2V A14B uses Video diffusion MoE transformer with 27B / 14B MoE. Pipeline and configuration fields are reported only where the official source publishes them.
Verified Video diffusion MoE transformer architecture
Official checkpoint and pipeline evidence
Official BF16, Unknown checkpoint coverage
Self-hosted Video generation workloads
Controlled model evaluation
Production deployment planning
Certified checkpoints and quantizations
Only exact, source-certified variants attached to this release are shown.
Recommended hardware
Use only the listed runtime configurations with the exact checkpoint precision. Hardware sizing considers resident pipeline memory, runtime overhead, input and output dimensions, batch size, and safety headroom.
Estimated weight memory51.8 GB
Runtime overhead23.48 GB
Total with safety margin93.35 GB
Recommended best value
2× NVIDIA A100 SXM 80GB
160 GB total VRAM · nvlink
66.65 GB headroom · Available
USD 1,830 / monthLowest-cost viable
4× NVIDIA L4 24GB
96 GB total VRAM · pcie
2.65 GB headroom · Available
USD 1,000 / monthHighest performance
2× NVIDIA A100 SXM 80GB
160 GB total VRAM · nvlink
66.65 GB headroom · Available
USD 1,830 / monthLicense and self-hosting considerations
The exact pipeline and selected precision must fit on the chosen GPUs. Requirements vary with resolution, duration, batch size, modalities, and runtime; language-model KV-cache rules are not applied.
Wan2.2 T2V A14B is published under Apache License 2.0. The release-scoped official license link and verification date are retained with the catalogue record.
Commercial Use Allowed. The official license permits commercial use; its obligations still apply.
Alternatives and sibling variants
Wan2.2 TI2V 5B, Wan2.2 I2V A14B, Wan2.2 S2V 14B, Wan2.2 Animate 2 14B
Release timeline
wan-ai/wan/wan2-2-t2v-a14b
Source details
Source revision: b55e10697e2ea2c5cbb86dbcd4f494c8cc51e3ab49e8da56d5cea79db793da9b
Frequently asked questions
How much VRAM does Wan2.2 T2V A14B require?
The estimate uses the certified checkpoint, precision, runtime, workload dimensions, batch size, and domain-specific overhead.
Which runtimes support Wan2.2 T2V A14B?
Certified upstream runtime evidence is available for wan, diffusers.
Can Wan2.2 T2V A14B be used commercially?
The stored commercial classification is Commercial Use Allowed. Consult the official Apache License 2.0 terms for the final obligations.
Official references
Last verified: 2026-08-26