Commit History

  1. Card: the root export is gone from main (revision 216f2841 keeps it)

    Verified signature
    fd7bb96
  2. Remove the 2026-07-20 root export (int4 + edit transformers): no app reads it since 2026-10-10; it stays at revision 216f2841

    Verified signature
    e854549
  3. Card: the zoo app moved to macos-int8 / macos-fp16 on 2026-10-10; the root files are no longer read by any app

    Verified signature
    216f284
  4. Card: the 2026-10-10 re-export (macos-fp16, macos-int8), measured table and component gate, text encoder corrected to the in-repo Qwen3 (4B shape)

    Verified signature
    039db98
  5. Add macos-int8/: re-export with apple/coreai-models 97a14be (gate vs fp32 PASS; see MANIFEST.json)

    Verified signature
    6324d15
  6. Add macos-fp16/: re-export with apple/coreai-models 97a14be (gate vs fp32 PASS; see MANIFEST.json)

    Verified signature
    1ac3254

  1. Card: note that this export does not load on current apple/coreai-models main (metadata_version); re-export planned

    Verified signature
    2f96f3c

  1. Card: macOS/iOS 27 GA wording (beta requirement dropped; beta findings dated)

    Verified signature
    311104a

  1. Card: DeviceMark row (2026-09)

    Verified signature
    bc77c01

  1. Card: definition line (Core AI, 2026-09)

    Verified signature
    f616ea0

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

    Verified signature
    e622639

  1. Link the card back to its collection and the request box

    Verified signature
    74ee807

  1. Add config.json so the Hub can count downloads

    Verified signature
    277fe5c

  1. strip_debug_info rescue: 0.4.0 debug locations removed (fixes OS 27 beta 2+ load), weights/graph unchanged

    Verified signature
    24782de

  1. Re-export with coreai-torch 0.4.1 (fixes OS 27 beta 2+ load failure)

    Verified signature
    76e080e
  2. Re-export with coreai-torch 0.4.1 (fixes OS 27 beta 2+ load failure)

    Verified signature
    5c6300d
  3. Re-export with coreai-torch 0.4.1 (fixes OS 27 beta 2+ load failure)

    Verified signature
    10fe099
  4. Re-export with coreai-torch 0.4.1 (fixes OS 27 beta 2+ load failure)

    Verified signature
    bf83fcb
  5. Re-export with coreai-torch 0.4.1 (fixes OS 27 beta 2+ load failure)

    Verified signature
    12b4952
  6. Re-export with coreai-torch 0.4.1 (fixes OS 27 beta 2+ load failure)

    Verified signature
    86b99dd

  1. Document multi-reference in-context editing

    Verified signature
    7a2eaed
  2. Add 2-reference in-context edit transformer (seq 12288)

    Verified signature
    2a4daec
  3. Document in-context editing (Transformer_edit)

    Verified signature
    54e5878
  4. Add in-context edit transformer (512, seq 2048)

    Verified signature
    6c37cea
  5. Add in-context edit transformer (1024, seq 8192)

    Verified signature
    d7b2038

  1. Card: mark macOS-only (4B over iPhone memory limit); fix app link to zoo apps/CoreAIImageGen

    Verified signature
    592d3ab
  2. Remove iOS components (Transformer_512 / VAE *_half): 4B exceeds iPhone memory; macOS-only

    Verified signature
    235f287
  3. Add v0.2 metadata.json descriptor (required by current CoreAIDiffusionPipeline runtime)

    Verified signature
    83a525a
  4. add iOS (512/half) component VAEEncoder_half.aimodel

    Verified signature
    8e860ff
  5. add iOS (512/half) component VAEDecoder_half.aimodel

    Verified signature
    fd14e0b
  6. add iOS (512/half) component Transformer_512.aimodel

    Verified signature
    5f17745
  7. FLUX.2 klein 4B: Core AI bundle (macOS, 1024, 4bit)

    Verified signature
    64ad4f8
  8. FLUX.2 klein 4B: Core AI model card

    Verified signature
    9d64097
  9. initial commit

    Verified signature
    fcc6f00