Card: DeviceMark row (2026-09)
Browse files
README.md
CHANGED
|
@@ -22,6 +22,10 @@ base_model_relation: quantized
|
|
| 22 |
|
| 23 |
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).
|
| 24 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 25 |
# BitCPM-8B → Apple Core AI (1.58-bit ternary, on-device)
|
| 26 |
|
| 27 |
The zoo's **first 1.58-bit ternary LLM** and **first sub-int8 packed-GEMM Metal kernel**, running
|
|
|
|
| 22 |
|
| 23 |
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).
|
| 24 |
|
| 25 |
+
<!-- gen-cards:devicemark begin (managed by scripts/gen-cards + tools/devicemark_row.py — edit cards.json, not this block) -->
|
| 26 |
+
This model has no row on [DeviceMark](https://devicemark.github.io/), the on-device LLM leaderboard.
|
| 27 |
+
<!-- gen-cards:devicemark end -->
|
| 28 |
+
|
| 29 |
# BitCPM-8B → Apple Core AI (1.58-bit ternary, on-device)
|
| 30 |
|
| 31 |
The zoo's **first 1.58-bit ternary LLM** and **first sub-int8 packed-GEMM Metal kernel**, running
|