Commit History

18 days ago

  1. Card: one Engine per conversation on LiteRT-LM (own report #3165), measured on litert-lm 0.18.0

    Verified signature
    04cacdd

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

    Verified signature
    7e6e081

  1. Card: measured device / browser results (edge-compat, 2026-09)

    Verified signature
    0fc6e8c

  1. Card: base_model_relation: quantized (list under the base model's Quantizations)

    Verified signature
    3310f6c

  1. Add measured Raspberry Pi 5 (CPU) section⁠

    Verified signature
    97565fa

  1. Card: correct the start_token note — measured numbers do not carry over

    Verified signature
    695c6a5
  2. Card: note the start_token fix (weights unchanged)

    Verified signature
    7b41085
  3. Drop the metadata start_token (weights unchanged)⁠

    Verified signature
    d5f2dc5

  1. Add measured Galaxy S26 GPU backend section⁠

    Verified signature
    aab72d6

  1. Zamba2-2.7B int8 (fp32act, full prefill ladder)

    Verified signature
    bf050bb
  2. Zamba2-2.7B LiteRT-LM card

    Verified signature
    436d4d1
  3. initial commit

    Verified signature
    a4ae3e7