anemll's picture
Upload folder using huggingface_hub
3d6aecb verified
model_info:
name: anemll-Qwen-Qwen3-0.6B-ctx512
version: 0.3.4
description: |
Demonstarates running Qwen-Qwen3-0.6B on Apple Neural Engine
Context length: 512
Batch size: 64
Chunks: 1
license: MIT
author: Anemll
framework: Core ML
language: Python
architecture: qwen3
parameters:
context_length: 512
batch_size: 64
lut_embeddings: none
lut_ffn: 6
lut_lmhead: 6
num_chunks: 1
model_prefix: qwen
embeddings: qwen_embeddings.mlmodelc
lm_head: qwen_lm_head_lut6.mlmodelc
ffn: qwen_FFN_PF_lut6.mlmodelc
split_lm_head: 16