CivArchive
    Wan 2.2 – SVI Pro with LoRA - v1.0b
    NSFW

    This is a workflow for Wan 2.2 SVI Pro with LoRa. It has a "loop" which allows you to set the number of passes without having to restructure the workflow. Only the section with the LoRa files needs to be expanded if you need more than one LoRa file per section or a higher total number of LoRa files.

    Model Links:

    https://huggingface.co/Kijai/WanVideo_comfy/blob/main/LoRAs/Stable-Video-Infinity/v2.0/SVI_v2_PRO_Wan2.2-I2V-A14B_HIGH_lora_rank_128_fp16.safetensors

    https://huggingface.co/Kijai/WanVideo_comfy/blob/main/LoRAs/Stable-Video-Infinity/v2.0/SVI_v2_PRO_Wan2.2-I2V-A14B_LOW_lora_rank_128_fp16.safetensors

    Used Models:

    https://civarchive.com/models/2190659/dasiwa-wan-22-i2v-14b-tastysin-v8-or-lightspeed-or-gguf

    Alternatively:

    https://huggingface.co/jayn7/WAN2.2-I2V_A14B-DISTILL-LIGHTX2V-4STEP-GGUF/tree/main

    https://huggingface.co/jayn7/WAN2.2-I2V_A14B-DISTILL-LIGHTX2V-4STEP-GGUF/tree/main/high_noise_1030?show_file_info=high_noise_1030%2Fwan2.2_i2v_A14b_high_noise_lightx2v_4step_1030-Q6_K.gguf

    https://huggingface.co/jayn7/WAN2.2-I2V_A14B-DISTILL-LIGHTX2V-4STEP-GGUF/tree/main/low_noise?show_file_info=low_noise%2Fwan2.2_i2v_A14b_low_noise_lightx2v_4step-Q6_K.gguf

    Setup:

    1. Do your main settings

    2. Write your prompts

    3. Select your loRas (if needed)

    Go for it!

    Info:

    This workflow is configured for 16 GB of VRAM. If you have less, increase the "block_to_swap" value or use a higher quantization model.

    Description

    v1.0 had some incorrect connections with LoRa nodes.

    FAQ

    Comments (11)

    FferrettJan 11, 2026· 1 reaction
    CivitAI

    Well done

    blusztaks177Jan 11, 2026
    CivitAI

    Hi,

    I'm trying to use your workflow but I cannot see a node to load VEA.

    Thalion7
    Author
    Jan 11, 2026· 1 reaction

    The VAE node is collapsed. Click right on the node and select unpin. Click right on the node again and select expand. Now you can select your VAE.

    citywalker1127821Jan 11, 2026
    CivitAI

    Can your workflow support multiple prompts (like 1, 2, 3, 4, 5 separate prompts), with each one able to use its own respective LoRA and duration?One more question: Does your workflow support the standard/regular A14 Wan2.2 FP8 model?

    Thalion7
    Author
    Jan 11, 2026

    Yes and no, because you need the Wan 2.2 SVI model. There is also one for fp8.

    FferrettJan 11, 2026
    CivitAI

    fyi: I've had significantly better I2V consistency and camera control with darksidewalkers "TastySin" q8 gguf with this workflow.

    Thalion7
    Author
    Jan 11, 2026

    I found a problem in my workflow that decrease the output quality. I will update it soon.

    srsparky31956Jan 11, 2026
    CivitAI

    The first generation comes out poorly, but everything after in the loop comes out fine. Any thoughts what this might be?

    Thalion7
    Author
    Jan 11, 2026· 1 reaction

    That's funny. This workflow (v1.0) has a problem with decreasing quality in the loop. I found the problem and will post a new version soon

    CowgaryJan 12, 2026
    CivitAI

    Thanks for the WF. I tried it but found it's not looping correctly and it's not saving the video. I didn't touch the settings at video combine so it's using the default setting. As for the loops, I set it to 2 but I only got a 4 second video as the output Asset that isn't saved.

    Thalion7
    Author
    Jan 12, 2026

    It's not for looping. It's just a loop for the passes. So you don't have to rebuild your workflow. But i found a problem that the quality will decrease. I fixed it and will post a new one soon. But i want to do some more testing before i post it.

    Workflows
    Wan Video 2.2 I2V-A14B

    Details

    Downloads
    0
    Platform
    CivitAI
    Platform Status
    Deleted
    Created
    1/11/2026
    Updated
    4/27/2026
    Deleted
    1/13/2026

    Files

    wan22SVIProWithLora_v10b.zip

    Mirrors

    HuggingFace (1 mirrors)
    CivitAI (1 mirrors)