mlboydaisuke commited on
Commit
139b086
·
verified ·
1 Parent(s): fb3df19

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

Browse files
Files changed (1) hide show
  1. README.md +2 -0
README.md CHANGED
@@ -13,6 +13,8 @@ base_model: zai-org/GLM-OCR
13
  base_model_relation: quantized
14
  ---
15
 
 
 
16
  # GLM-OCR → Core AI
17
 
18
  On-device document OCR, running entirely on Apple's **Core AI** (Neural Engine / GPU).
 
13
  base_model_relation: quantized
14
  ---
15
 
16
+ Core AI is Apple's on-device ML runtime in iOS 27 / macOS 27 and the successor to Core ML: PyTorch models are exported with Apple's `coreai-torch` (LLMs: `coreai.llm.export`) into `.aimodel` bundles that run on the GPU or the Neural Engine, e.g. Qwen3-8B 4-bit decodes at 94 tok/s on an M4 Max GPU, MLX 90 under the same protocol ([apple-silicon-llm-bench](https://github.com/john-rocky/apple-silicon-llm-bench), macOS 27 beta, 2026-06).
17
+
18
  # GLM-OCR → Core AI
19
 
20
  On-device document OCR, running entirely on Apple's **Core AI** (Neural Engine / GPU).