veizik WAN (CUDA) v1.2.8 — Wan 2.1 & Wan 2.2 video — cuda-wan-1.2.8

Text-to-video and image-to-video with Wan2.1-T2V-1.3B, Wan2.1-T2V-14B, Wan2.2-T2V-A14B and Wan2.2-I2V-A14B on your own NVIDIA GPU. Veizik runs the models through its own native engine: no Python, no CUDA toolkit. You supply the weights; the frames are written as PNG files to the folder you name.. It also runs a resident render server with a host-loop (bin/veizik-resident) that keeps one process and one GPU context across requests, and mid-render preemption — an urgent request interrupts a running render, is served, then the first render resumes.

This page records the state as of 2026-10-28 and is not updated as the catalogue moves. For what runs now, see models.

This build has been superseded and is no longer distributed. The digest below still identifies it, so a copy you already hold can be checked against it. The current build is cuda-wan-1.2.93.

sha256 3fbe53c3bedb9158b00b0e1ab9a605af58a9ed5d201cefd2c0bad717ebc3d047

What this build does

• veizik run <model key> "prompt" --out DIR renders a video and writes its frames as PNG files to DIR/frames/; nothing is written inside the package.
• Serve: veizik serve <model> runs a resident render server that renders request after request in one process and one GPU context, with results bit-identical to veizik run.
• Four modes: quality (the reference path and the default), balanced, fast and ultrafast.
• An interrupted render continues from its last step with --resume, and the result is identical to a render that never stopped.
• Every file of the package is checked against its sealed manifest before each run.
• GPU code for sm_80, sm_86, sm_89, sm_90a, sm_100a and sm_120a.
• Every frame it writes carries the provenance mark described in the Terms: an in-pixel watermark plus XMP DigitalSourceType trainedAlgorithmicMedia.
• Serve: veizik serve wan22_t2v_a14b --port 8000 is a render API on 127.0.0.1 (POST /v1/videos/generations with a prompt and seed returns the folder of PNG frames).
• veizik serve runs a resident render server (bin/veizik-resident) — the model loads once and is reused across requests with no reload.
• mid-render preemption — POST /v1/videos/generations with "priority": "urgent" interrupts a running render at a step boundary, is served, then the first render resumes from its checkpoint.

Models in this release

Platform

Linux on x86-64

Verification

`veizik install` checks the download against SHA-256 3fbe53c3bedb9158b00b0e1ab9a605af58a9ed5d201cefd2c0bad717ebc3d047 and refuses any other file. The release is reproducible: SHA256SUMS on the release, a single-file .vzpack with a signed .vzindex. To check veizik-wan-1.2.8-linux-x86_64.vzpack by hand, compare it with sha256sum.

File

veizik-wan-1.2.8-linux-x86_64.vzpack — no longer downloadable

Size

6,851,149 bytes

SHA-256

3fbe53c3bedb9158b00b0e1ab9a605af58a9ed5d201cefd2c0bad717ebc3d047

GPU architectures

sm_80, sm_86, sm_89, sm_90a, sm_100a, sm_120a

Run end to end on

GeForce RTX 3090 (sm_86); GeForce RTX 3090 (sm_86); GeForce RTX 3090 (sm_86)

Release

v1.2.8

Get, install, activate

# the veizik command for Linux x86-64
$ curl -fsSL https://veizik.com/install.sh | sh
# if the installer asks you to put ~/.local/bin on your PATH
$ export PATH="$HOME/.local/bin:$PATH"
# accepted names: wan21, wan22 or a full model key (wan21_t2v_1_3b, wan21_t2v_14b, wan22_t2v_a14b, wan22_i2v_a14b); any case; '.' and '-' work too (wan2.2)
$ veizik install wan22
$ veizik activate

Run `veizik activate` once per device after `veizik install`. Then `veizik pull wan22` fetches the Wan weights and prepares the video decoder for you — no manual VAE setup or support request. Activation itself does not need the weights; rendering does. veizik install refuses a download whose SHA-256 differs from this release. The other packages are on the download page.

Fixed issues

• Resident and long renders now resume from their saved checkpoint and continue the interrupted render — byte-identical to a render that never stopped — instead of restarting it (the resident host-loop forwards --resume).
• Checkpoints saved by a long or resident render now use a consistent render.vzckpt filename. Renders and output frames are byte-identical to 1.2.5 (engine and GPU code unchanged).
• Maintenance re-cut: the same four models and byte-identical renders as 1.2.3, with the 1.2.4 install-reliability improvements (clean archive, version-stable unpack directory, CUDA 13.x lease provisioning) carried forward.
• Online device activation works again.
• The package no longer includes macOS AppleDouble (._*) files.
• An internal verification step no longer stops a run on a benign mismatch.

Known issues in cuda-wan-1.2.8

• Model weights are not included. veizik pull wan22 downloads the Wan2.1/2.2 weights (Apache-2.0) and prepares the video decoder automatically — no manual VAE setup; veizik pull --prep-local DIR prepares weights you already have.
• Speed figures for this package were measured on an NVIDIA RTX 3090 (sm_86) only.
• The image-to-video model takes its input image as raw RGB bytes, one byte per channel, at exactly the render size (width × height × 3 bytes).
• balanced, fast and ultrafast are not bit-identical to quality; without --det-vae yes decoded frames may differ slightly from run to run.