Qwen-Image-Layered
Official · BF16 · safetensors
Runtime certified: Diffusers
| Precision / quantization | Format | Bits | Publisher trust | Deployment state |
|---|---|---|---|---|
| BF16 | safetensors | 16 | Official | Runtime certified |
Qwen Image · Layered image · Qwen Image Layered
Qwen Image Layered: Specs, GPU requirements, hosting & deployment
Qwen Image Layered is a current Layered Image release in the Qwen Image family, certified from its immutable official source revision.
Total parameters20.4B
Active parametersDense model
Context windowN/A / Not directly comparable
Release statusCurrent · Current
ArchitectureLayer-decomposition diffusion transformer
ModalitiesText
Licenseapache-2.0
Deployment stateDeployable with certified configuration
Commercial stateCommercial Use Allowed
Release date2025-12-17
Architecture, strengths and use cases
Qwen Image Layered is an official self-hostable Qwen release in the Qwen Image Layered Image track. It uses Layer-decomposition diffusion transformer, carries 20.43B, supports a verified context window of 0 tokens, and exposes only source-certified checkpoints and runtime combinations. Its release, license and deployment facts are tied to the official repository revision rather than inferred from the family name.
Qwen Image Layered uses Layer-decomposition diffusion transformer with 20.43B. The stored architecture profile records layers, hidden width , attention heads and a 0 token context ceiling where the official configuration publishes those fields.
Verified Layer-decomposition diffusion transformer architecture
Official checkpoint and pipeline evidence
Official BF16, Unknown checkpoint coverage
Private LLM inference
Agentic workflows
Long-context document processing
Certified checkpoints and quantizations
Only exact, source-certified variants attached to this release are shown.
Recommended hardware
Use only the listed diffusers runtime configuration with the exact checkpoint precision. The recommendation engine checks weight memory, KV cache, runtime overhead, safety headroom, GPU architecture and topology before ranking rentable and reference hardware.
Estimated weight memory39.2 GB
KV cache0 GB
Runtime overhead16.5 GB
Total with safety margin69.06 GB
Recommended best value
1× NVIDIA H200 SXM 141GB
141 GB total VRAM · nvswitch
71.94 GB headroom · Limited capacity
USD 2,345 / monthLowest-cost viable
3× NVIDIA L4 24GB
72 GB total VRAM · pcie
2.94 GB headroom · Available
USD 750 / monthHighest performance
1× NVIDIA H200 SXM 141GB
141 GB total VRAM · nvswitch
71.94 GB headroom · Limited capacity
USD 2,345 / monthLicense and self-hosting considerations
The full 20.43B resident weight set must fit across the selected GPUs. Context and concurrency add KV-cache memory; MoE active parameters do not replace resident-weight memory.
Qwen Image Layered is published under apache-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
Qwen-Image-Layered
Release timeline
qwen/qwen-image/qwen-image-layered
Source details
Source revision: 881ef920eca3fd52232cd80df5d6bd2d954d894654e67fb5ab051beb30615985
Frequently asked questions
How much VRAM does Qwen Image Layered require?
The exact estimate depends on checkpoint precision, context and concurrency; the page recommendation uses the certified memory engine and compatible runtime configurations.
Which runtimes support Qwen Image Layered?
Certified upstream runtime evidence is available for diffusers.
Can Qwen Image Layered be used commercially?
The stored commercial classification is Commercial Use Allowed. Consult the official apache-2.0 terms for the final obligations.
Official references
Last verified: 2026-09-03