Commit History

29 days ago

  1. docs: update model card, context lengths, and config specs

    8c97472

  1. update models

    58d9b7d

  1. Stop restating quantized, and drop the unused default⁠

    f084973
  2. Correct the published metadata⁠

    9313777

  1. Re-export the 8da4w text variants at the sweep's group size (#1)⁠

    191e308

  1. Update chat template for strict monotonicity and tool support

    ab65e00

  1. Use min/max dynamic shape bounds for forward inputs/outputs

    08c22ab
  2. Backfill constant metadata method values in config.json from .pte

    3df529f

  1. Rename 1_2b/mlx/lfm2_5_1_2b_mlx_4w_v2.pte to 1_2b/mlx/lfm_2_5_1_2b_mlx_int4.pte

    Verified signature
    07921ad
  2. Upload lfm2_5_1_2b_mlx_4w_v2.pte

    Verified signature
    26bf6f8
  3. Create mlx/config.json

    Verified signature
    66ee684

  1. Add spec-compliant config.json files

    Verified signature
    a442f70
  2. Fix lfm-2.5 layout: distinguish VL from text + colocate tokenizers

    Verified signature
    9e10ff2
  3. Add spec-compliant config.json files

    Verified signature
    3316099
  4. Restructure to MODEL_SPEC.md convention

    Verified signature
    46260eb