Instructions to use mochiexists528/ace-step-mlx-planner-1.7b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use mochiexists528/ace-step-mlx-planner-1.7b with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir ace-step-mlx-planner-1.7b mochiexists528/ace-step-mlx-planner-1.7b
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
Configuration Parsing Warning:Config file tokenizer_config.json cannot be fetched (too big)
ACE-Step MLX Planner 1.7B (fp16)
MLX-compatible fp16 package for the ACE-Step 5Hz planner LM used by optional Phase 6 LM-conditioning. This repo is shared by the base turbo and XL-turbo ACE-Step MLX model packages so the planner is not duplicated in every DiT repo.
Repository contents
config.json
model.safetensors
tokenizer.json
tokenizer_config.json
special_tokens_map.json
added_tokens.json
chat_template.jinja
merges.txt
vocab.json
Usage
Baseline generation does not need this repo. To enable Phase 6, pass the
planner snapshot to the generator with --lm-weights.
For Q4 DiT variants, also pass the model repo's bundled
detokenizer.safetensors as --lm-conditioning-checkpoint. For fp16 DiT
variants, the FSQ projection and AudioTokenDetokenizer tensors are inline in
dit/model.safetensors, so no detokenizer sidecar is needed.
hf download mochiexists528/ace-step-mlx-planner-1.7b --local-dir ./planner-fp16
Source attribution
The planner LM is the ACE-Step 5Hz planner, a fine-tune in the Qwen3-1.7B family. Qwen3 is Apache 2.0 licensed. This package changes only the on-disk layout and MLX-compatible safetensors dtype representation; it does not change the model architecture or learned parameters.
NOTICE
This package carries the upstream Qwen3 NOTICE obligation. See the upstream Qwen3 model card for the canonical NOTICE text and verify the latest version before re-redistribution.
- Downloads last month
- 25
Quantized