Hugging Face's logo

mlboydaisuke
/
TwIL-LM3-LiteRT

Text Generation
LiteRT-LM
LiteRT
litertlm
on-device
edge
Model card Files
xet
 Community
TwIL-LM3-LiteRT

Commit History

3 days ago

  1. Thinking cannot be switched off, and the think tags come back inside the text: litert-lm 0.18.0 (measured 2026-10-13)

    mlboydaisuke
    Verified signature
    7159d19

25 days ago

  1. Chat template: accept the 0.18 content-parts form (string form unchanged); weights, tokenizer and executor metadata byte-identical

    mlboydaisuke
    Verified signature
    210bbfc

  1. Card: base_model_relation: quantized (so the conversion lists under the base model's Quantizations, not Finetunes)

    mlboydaisuke
    Verified signature
    d89b07d

  1. Card: restore the section dropped by the 2026-08 note script (text after 'Conversion notes')

    mlboydaisuke
    Verified signature
    b42daf0

  1. Card: note the default-system-prompt fix (weights unchanged)

    mlboydaisuke
    Verified signature
    4210f37

  2. Restore the vendor's default system prompt in the chat template (weights unchanged). The converter's template probe never renders the block the upstream template emits when the caller sends no system message, so the bundle silently dropped it; with no system message the prompt now renders byte-identical to the upstream chat template's output (see the card note).

    mlboydaisuke
    Verified signature
    9f56307

  1. card: the formal-logic A/B against the base - 10/12 vs 9/12, no difference detected

    mlboydaisuke
    Verified signature
    101b369

  2. TwIL-LM3 int4 (blockwise-32 + OCTAV, externalised embedder)

    mlboydaisuke
    Verified signature
    0191fc0

  3. TwIL-LM3 LiteRT-LM card

    mlboydaisuke
    Verified signature
    6e448fc

  4. webAI Non-Commercial License v1.0 (required by its SS3.1)

    mlboydaisuke
    Verified signature
    82095b3

  5. initial commit

    mlboydaisuke
    Verified signature
    617ba42