Hugging Face's logo

mlboydaisuke
/
LFM2.5-1.2B-CoreAI

Text Generation
coreai
coreai-aimodel
aimodel
apple-silicon
on-device
lfm2
hybrid
Model card Files
xet
 Community
LFM2.5-1.2B-CoreAI / gpu-pipelined

Commit History

  1. lfm2_5_1_2b_instruct_decode_int8lin: re-save with coreai-core 1.0.0b2 (strip debug locations; weights unchanged). The 0.4.0-era IR refused AIModel.load on every OS 27 build since beta 2, still on the RC 26A428 (2026-09-14); the re-saved bundle loads. Recovery: coreai-model-zoo conversion/recovery/{strip_b1,resave_b2}.py

    mlboydaisuke
    Verified signature
    eaf9579

  1. Re-export with coreai-torch 0.4.1 (fixes OS 27 beta 2+ load failure); gated token-for-token vs fp32 oracle

    mlboydaisuke
    Verified signature
    8dc3742

  1. Upload folder using huggingface_hub

    mlboydaisuke
    Verified signature
    56b74ab

  1. gpu-pipelined int8lin bundle: 253 tok/s M4 Max / 38-39.6 iPhone 17 Pro, oracle gate 16/16

    mlboydaisuke
    Verified signature
    d67f006