Hugging Face's logo

Spaces:
build-small-hackathon
/
unstuck
Paused

App Files Community
unstuck / src

Commit History

  1. fix(recall): default embed model to Qwen3-Embedding-8B

    art87able
    Verified signature
    52b7250

  2. deploy: multi-backend (minicpm/nemotron/finetuned) + 3-tier ZeroGPU->serverless->local-finetune fallback; finetune pipeline

    art87able
    fe7f8fd

  3. feat: similar-task recall via Nebius embeddings (keyless; recall degrades off without NEBIUS_API_KEY)

    art87able
    Verified signature
    79e91df

  1. deploy: offgrid on-device backend

    art87able
    Verified signature
    f852d34

  1. Upload folder using huggingface_hub

    art87able
    Verified signature
    ce5b6f9

  2. Upload folder using huggingface_hub

    art87able
    Verified signature
    7a754d7

  3. Upload src/unstuck/store.py with huggingface_hub

    art87able
    Verified signature
    5338711

  4. Upload src/unstuck/prompts.py with huggingface_hub

    art87able
    Verified signature
    81858de

  5. Upload src/unstuck/store.py with huggingface_hub

    art87able
    Verified signature
    f9fc575

  6. Upload src/unstuck/prompts.py with huggingface_hub

    art87able
    Verified signature
    e75a19e

  1. fix: thread-safe SQLite store

    art87able
    Verified signature
    a98f11a

  2. fix: unpack BatchEncoding into model.generate

    art87able
    Verified signature
    3080b01

  3. fix(zerogpu): load model with .to(cuda) instead of device_map

    art87able
    Verified signature
    012de31

  4. Deploy Unstuck (Codex-built) — initial code

    art87able
    Verified signature
    2f62bd4