Hugging Face's logo Hugging Face
  • Models
  • Datasets
  • Spaces
  • Buckets new
  • Docs
  • Enterprise
  • Pricing
    • Website
      • Tasks
      • HuggingChat
      • Collections
      • Languages
      • Organizations
    • Community
      • Blog
      • Posts
      • Daily Papers
      • Hardware
      • Learn
      • Discord
      • Forum
      • GitHub
    • Solutions
      • Team & Enterprise
      • Hugging Face PRO
      • Enterprise Support
      • Inference Providers
      • Inference Endpoints
      • Storage Buckets

  • Log In
  • Sign Up

mlboydaisuke
/
granite-4.0-h-CoreAI

coreai
coreai-aimodel
apple
aimodel
on-device
granite
mamba2
apple-silicon
macos
ios
hybrid
granite-4.0-h
tokens-per-second
m4-max
iphone-17-pro
Model card Files Files and versions
xet
Community
granite-4.0-h-CoreAI / gpu-pipelined
4.27 GB
Ctrl+K
Ctrl+K
  • 1 contributor
History: 6 commits
mlboydaisuke's picture
mlboydaisuke
granite_4_0_h_350m_decode_fp16: 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
f361cfc verified about 1 month ago
  • granite_4_0_h_1b_decode_int8hu_block32_sym
    Re-export with coreai-torch 0.4.1 (fixes OS 27 beta 2+ load failure); gated token-for-token vs fp32 oracle 3 months ago
  • granite_4_0_h_1b_decode_int8lin
    Re-export with coreai-torch 0.4.1 (fixes OS 27 beta 2+ load failure); gated token-for-token vs fp32 oracle 3 months ago
  • granite_4_0_h_350m_decode_fp16
    granite_4_0_h_350m_decode_fp16: 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 about 1 month ago