Wan 2.1 · Wan 2.2

From the documents of veizik-wan-1.2.8-linux-x86_64.vzpack (release v1.2.8).

• Four video models in one package: Wan2.1-T2V-1.3B, Wan2.1-T2V-14B and Wan2.2-T2V-A14B for text to video, and Wan2.2-I2V-A14B for image to video.
• Run: veizik run <model key> "prompt" --out DIR renders a video and writes its frames as PNG files to DIR/frames/. Options: --steps, --frames, --width, --height, --seed, --guidance, --guidance2, --shift, --mode, --neg-file, --image, --det-vae, --checkpoint-every and --png-step; --lora and --lora-scale are accepted, but LoRA adapters are not part of this release. Any other option is refused with <flag> does not apply to this model. Every run also takes --out, --name, --prompt-file, --resume and --resume-from. Nothing is written inside the package.
• Serve: veizik serve wan22_t2v_a14b --port 8000 runs a resident render server — the model loads once and is reused across requests with no reload, in one process and one GPU context, and the resident server re-checks the licence on each request. POST /v1/videos/generations with {"prompt": "…", "seed": 42} renders and returns the folder of PNG frames; /health and /v1/models answer too. An optional "priority": "urgent" on a request interrupts a running render at a step boundary, saves its progress (.vzckpt), serves the urgent request, then resumes the first. A non-resident bundle renders one request at a time.
• Image to video: Wan2.2-I2V-A14B takes its first frame with --image FILE, as raw RGB bytes (one byte per channel) at the render size.
• Four modes: quality (the reference path and the default), balanced, fast and ultrafast. Only quality is bit-identical to the reference.
• Memory, per model: Wan2.1-T2V-1.3B runs in about 7 GB of VRAM; the 14B models (Wan2.2-T2V-A14B and Wan2.2-I2V-A14B) render resident in about 18.5 GB on a 24 GB consumer card (peak near 22 GB for T2V-A14B). Host system memory stays around 2 GB (peak near 1.83 GB at the VAE decode of an 81-frame clip) — the weights stream to the GPU rather than loading into system RAM, so a 14B video model runs on an ordinary machine.
• Resume: an interrupted render continues with --resume and the same --out (or --resume-from REF, where REF is a --name or an absolute run directory), and the result is bit-identical to a render that never stopped. Resume finishes the same render — same prompt, seed and settings — so to change the prompt, start a new render. --checkpoint-every N sets how often progress is saved.
• Preemption: create DIR/work/preempt and the render saves its state at the next step boundary and exits 75; resume it as above.
• Progress: DIR/work/progress always holds one JSON line (state, stage, step, window, frames committed, estimated time left), replaced atomically, so a program never has to parse logs.
• Deterministic VAE: with --det-vae yes the same input gives byte-exact frames.
• Provenance mark: every frame carries the v1 mark, an in-pixel watermark plus XMP DigitalSourceType trainedAlgorithmicMedia. It holds no prompt, device, path or account data.
• Kernel sets for sm_80, sm_86, sm_89, sm_90a, sm_100a and sm_120a; the launcher picks the set for the GPU it finds. Every speed figure for this package was measured on an RTX 3090 only.
• veizik install downloads the package from its public GitHub release and refuses it unless its SHA-256 matches the published release. Every file is checked against its sealed manifest before each run; a changed, added or missing file stops the run with exit code 4.
• Activation is veizik activate, once per device. veizik pull wan then fetches the Wan weights and prepares the video decoder automatically (no manual VAE layout step); for weights you already have, veizik pull --prep-local DIR prepares them in place.

Availability by variant

Verified on

• GeForce RTX 3090 · build cuda-wan-1.2.8 · 2026-10-28 · published measurement

Earlier-build evidence

On 2026-10-21 an earlier build of the Wan engine rendered all four models bit-exact against the package's own reference hashes (CONFORMANCE.json) on an NVIDIA RTX 3090; the shipped engine is a later rebuild and has not been re-rendered.

“Verified” means this family ran end to end on that exact device, on a published build. Architecture membership never upgrades a device to verified.

Each table shows the GPUs a model was measured on and one figure per measure: the default mode and the most common workload. Hover a figure for its build, date, runs and range; open All figures for every mode and workload. A figure marked earlier build is our own measurement of an earlier build of the engine, shown because the computation it exercises did not change in substance; it is replaced when the package you download is measured on that GPU.