Hugging Face's logo

mlx-community
/
clef-flash-4bit

Zero-Shot Classification
MLX
Safetensors
qwen3_5
clef
cloudflare
systemone
structured-output
classification
multimodal
custom-code
decision-model
Eval Results
4-bit precision
Model card Files
xet
 Community
2
clef-flash-4bit / clef_mlx.py

Commit History

22 days ago

  1. clef_mlx.py: use the repo name (not the cache snapshot hash) as the default model name

    lucataco
    Verified signature
    140bf7e

  2. clef_mlx.py: default --model works inside Hub cache snapshots (symlinked files)

    lucataco
    Verified signature
    686174e

  3. clef_mlx.py: add 'predict' CLI and 'serve' (local POST /v1/systemone server); card: CLI and server docs

    lucataco
    Verified signature
    7c95239

  4. Card: zero-shot-classification tag, RAM guide for all variants, limitations, pinned mlx-vlm; clef_mlx: warn on untested mlx-vlm/mlx-lm versions

    lucataco
    Verified signature
    6822f0f

  5. Add Decision Index quality check; clef_mlx: strict no-truncation mode (truncate=False)

    lucataco
    Verified signature
    d9ec324

23 days ago

  1. Add MLX 4-bit conversion of Cloudflare/clef-flash with joint schema head and clef_mlx.py loader

    lucataco
    Verified signature
    7cec35c