Hugging Face's logo

Spaces:
dagloop5
/
TestingRef2va
Running on Zero

App Files Community
TestingRef2va / app.py

Commit History

  1. Upload app.py

    amisima
    Verified signature
    0f38077

  2. Upload app.py

    amisima
    Verified signature
    4762142

  3. Upload app.py

    amisima
    Verified signature
    d0795e0

  1. Open badge links in a new tab

    multimodalart HF Staff
    Verified signature
    3320577

  2. Price bookings against the AoTI blocks, generate from 2 s again, cross-link the demos⁠

    multimodalart HF Staff
    92693c1

  3. Start the duration slider at the 5 s the model actually generates

    multimodalart HF Staff
    Verified signature
    74a631a

  4. Restore probe

    multimodalart HF Staff
    Verified signature
    97142dd

  5. Cache the conditioner client with functools.cache, tidy comments

    multimodalart HF Staff
    Verified signature
    1f48523

  6. Let gradio_client forward the caller's ZeroGPU token itself instead of pinning one by hand

    multimodalart HF Staff
    Verified signature
    b2bc645

  7. Forward the caller's ZeroGPU token to the conditioner, plainly

    multimodalart HF Staff
    Verified signature
    afea884

  8. Bill the conditioner call to the requesting user's own ZeroGPU token, never an org token

    multimodalart HF Staff
    Verified signature
    a4b3082

  9. Pay for the conditioner call with the first ZeroGPU identity that can: the caller's token, this Space's HF_TOKEN, then none

    multimodalart HF Staff
    Verified signature
    9539b35

  10. Fall back to an anonymous conditioner call when the forwarded ZeroGPU token is refused

    multimodalart HF Staff
    Verified signature
    e0f6bde

  11. Forward the caller's ZeroGPU token to the conditioner so one request bills as one request

    multimodalart HF Staff
    Verified signature
    e76d22a

  12. Load the AoTI packages from the public multimodalart/minimax-h3-aoti, move the upsample toggle under the prompt

    multimodalart HF Staff
    Verified signature
    4218806

  13. Migrate to canonical diffusers PR 14371 + public MiniMaxAI/MiniMax-H3 weights, add prompt upsampling toggle

    multimodalart HF Staff
    Verified signature
    705b884

  1. Duration slider from 2s

    multimodalart HF Staff
    Verified signature
    d9f8e44

  2. Mirror the conditioner's canonical table, grouped by ratio

    multimodalart HF Staff
    Verified signature
    6fe31a4

  3. Mirror the conditioner's 16-canvas table (fast 4:3, 3:4 and 21:9 added)

    multimodalart HF Staff
    Verified signature
    2225cd2

  4. Point at the public multimodalart/qwen3vl-conditioner, called without a token

    multimodalart HF Staff
    Verified signature
    1f02c70

  5. Fast canvas stays the default; the examples carry full-resolution canvases

    multimodalart HF Staff
    Verified signature
    1aa28b5

  6. Full 16:9 as the default canvas; match the conditioner's renamed fast label

    multimodalart HF Staff
    Verified signature
    6237c6a

  7. Dynamic GPU duration from the packed sequence, cached examples, torchaudio note

    multimodalart HF Staff
    Verified signature
    f03920f

  8. torchaudio for off-rate reference soundtracks; title, fillable width, slot height

    multimodalart HF Staff
    Verified signature
    4bdd7d9

  9. Images/audio/video tab order, expandable image slots, native attention for the VAEs

    multimodalart HF Staff
    Verified signature
    79e1481

  10. Mirror the generator's title, theme and advanced-options layout

    multimodalart HF Staff
    Verified signature
    cf77190

  11. One tab per reference modality, in reading order

    multimodalart HF Staff
    Verified signature
    508462c

  12. MiniMax-H3 ref2va, the denoising half of the split deployment

    multimodalart HF Staff
    Verified signature
    9e3b8ca