The model in this repository is an MLX model converted from the pre-trained J-Moshi model for Mac OS. For other terms and restrictions, it adheres to the original models nu-dialogue/j-moshi-ext and kyutai/moshi.

このリポジトリのモデルは,J-Moshiの学習済みモデルをMac OS向けに変換したMLXモデルです。 その他規約・制限等においてはオリジナルのモデルであるnu-dialogue/j-moshi-ext並びにkyutai/moshiに準拠します。

Installation

pip install moshi_mlx

詳細はhttps://github.com/kyutai-labs/moshi?tab=readme-ov-file#requirements を参照。

Example Usage

python -m moshi_mlx.local --hf-repo akkikiki/j-moshi-ext-mlx
Downloads last month
939
Safetensors
Model size
7.69B params
Tensor type
BF16
·
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API: The model has no pipeline_tag.

Model tree for akkikiki/j-moshi-ext-mlx

Finetuned
(1)
this model
Quantizations
2 models