Hugging Face's logo

JohnP1
/
d1a-e2b-mlx-q8

Text Classification
MLX
Safetensors
English
Japanese
gemma4
d1a
decision-model
pull-requests
apple-silicon
quantized
on-device
8-bit precision
Model card Files
xet
 Community
d1a-e2b-mlx-q8

Commit History

4 days ago

  1. config.json: vision_config

    JohnP1
    Verified signature
    ae010a2

  2. Model card: wording

    JohnP1
    Verified signature
    859d573

16 days ago

  1. v0.6.1: routing temperature (use_case_temperatures.routing), same weights as v0.6 (jonpol01/d1a#264)

    JohnP1
    Verified signature
    190fca2

17 days ago

  1. Card: the chosen D1A avatar

    JohnP1
    Verified signature
    8af5915

  2. Card: add the D1A avatar

    JohnP1
    Verified signature
    b90e1bf

18 days ago

  1. Card: say what D1A is first, with an example

    JohnP1
    Verified signature
    b452545

19 days ago

  1. D1A-E2B v0.6

    JohnP1
    Verified signature
    f9479f2

26 days ago

  1. v0.2: add the base's vision and audio encoders (media/) for photos and voice

    JohnP1
    Verified signature
    841ba19

28 days ago

  1. Calibration temperature 1.52 (fitted on the study's calibration split; the published head.pt had been left at 1.0)

    JohnP1
    Verified signature
    438e045

  2. D1A-E2B v0.2 (two epochs) MLX q8, parity-checked

    JohnP1
    Verified signature
    044d7e9

29 days ago

  1. D1A-E2B v0.1 MLX q8 (per-layer embeddings 4-bit), parity-checked

    JohnP1
    Verified signature
    dd7f1e7

  2. initial commit

    JohnP1
    Verified signature
    d01fcc7