Hugging Face's logo

Spaces:
build-small-hackathon
/
one-for-all
Paused

App Files Community
one-for-all

Commit History

  1. raise ask cap 512->768 tokens (still ~30s, under ZeroGPU 60s cap)

    Kaiky Maia
    Verified signature
    a80fa48

  2. revert GPU duration to default (free ZeroGPU caps <120s); keep 512-token cap

    Kaiky Maia
    Verified signature
    0f66bac

  3. raise generation cap (200→512 ask, 160→384 arena) + GPU duration 120s

    Kaiky Maia
    Verified signature
    1e55370

  4. raise generation cap (200→512 ask, 160→384 arena) + GPU duration 120s

    Kaiky Maia
    Verified signature
    b7e4f73

  5. resilient umap transform (point=None fallback) + pin umap-learn 0.5.12

    Kaiky Maia
    Verified signature
    62736c4

  6. resilient umap transform (point=None fallback) + pin umap-learn 0.5.12

    Kaiky Maia
    Verified signature
    3fabf89

  7. fix device mismatch: move gating head to GPU alongside student

    Kaiky Maia
    Verified signature
    cf58bc4

  8. ZeroGPU: load student on CPU at startup with cuda.is_available guarded False

    Kaiky Maia
    Verified signature
    cd53ece

  9. ZeroGPU: load student on CPU at startup with cuda.is_available guarded False

    Kaiky Maia
    Verified signature
    680bfe9

  10. ZeroGPU: load student on CPU at startup (no device_map=auto), move to GPU in handler

    Kaiky Maia
    Verified signature
    2bd54d1

  11. back to torch backend + ZeroGPU; drop llama-cpp-python

    Kaiky Maia
    Verified signature
    7ec30d6

  12. fix n_threads oversubscription on cpu-basic (host reports 16+, container has 2)

    Kaiky Maia
    Verified signature
    8b7fcee

  13. llama.cpp speedups: n_threads, n_ctx 1024, gate off hot path, max_tokens 96

    Kaiky Maia
    Verified signature
    14f2b64

  14. llama.cpp backend: requirements.txt

    Kaiky Maia
    Verified signature
    a3e2f2f

  15. llama.cpp backend: _probe.py

    Kaiky Maia
    Verified signature
    059e92b

  16. fix deps: transformers>=5 (gradio 6.18 needs hub>=1.0)

    Kaiky Maia
    Verified signature
    907e59e

  17. drop stale __pycache__

    Kaiky Maia
    Verified signature
    82a5cc5

  18. Custom gr.Server frontend + 6-teacher model (Nemotron), token streaming, arena

    Kaiky Maia
    Verified signature
    9409aaa

  1. Upload folder using huggingface_hub

    Kaiky Maia
    Verified signature
    85bd34a

  2. Upload folder using huggingface_hub

    Kaiky Maia
    Verified signature
    0647956

  3. Upload folder using huggingface_hub

    Kaiky Maia
    Verified signature
    9a42152

  4. Upload folder using huggingface_hub

    Kaiky Maia
    Verified signature
    dbdd574

  5. Upload _html.py with huggingface_hub

    Kaiky Maia
    Verified signature
    5708a34

  6. Upload app.py with huggingface_hub

    Kaiky Maia
    Verified signature
    1050ae4

  7. Upload app.py with huggingface_hub

    Kaiky Maia
    Verified signature
    9ef5bac

  8. Upload app.py with huggingface_hub

    Kaiky Maia
    Verified signature
    44eb438

  9. Upload _glb.py with huggingface_hub

    Kaiky Maia
    Verified signature
    3d7ad32

  1. fix: gradio model 3d

    Kaiky Maia
    Verified signature
    703b5b1

  2. fix: iframe Three.js, model response output, UI labels

    Kaiky Maia
    Verified signature
    40b1944

  3. feat: Three.js 3D soul space (remove Plotly UMAP)

    Kaiky Maia
    Verified signature
    956241a

  4. fix: gating.pt path (root, not final/)

    Kaiky Maia
    Verified signature
    cc58dce

  5. fix: gradio 5.33.2 (python 3.13 compat)

    Kaiky Maia
    Verified signature
    d7ada6c

  6. Initial Space — OFA soul transfer

    Kaiky Maia
    Verified signature
    6ef56c8

  7. Initial Space — OFA soul transfer

    Kaiky Maia
    Verified signature
    275cef9

  8. initial commit

    Kaiky Maia
    Verified signature
    91cae6d