Downloads · 30 days
0
moshika-mlx-mp
strumecki/moshika-mlx-mp
moshika-mlx-mp is a machine learning model from strumecki. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for mlx. The card lists the license as cc-by-4.0.
Mixed-precision MLX checkpoint of Kyutai Moshika. Targets ~5 GB of weights, making the 7B Moshi practical on 8 GB Apple Silicon Macs while keeping the quality-critical layers at q8.
Access
Public
Updated May 16, 2026
Repo size
5.3 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors5.3 GB · 100%
At a glance
- Library
- mlx
- License
- cc-by-4.0
- Access
- Public
- Created
- May 16, 2026
- Updated
- May 16, 2026
- SHA
- be8296ad
Base models
Model card
From the Hugging Face model README
Moshika MLX — Mixed Precision (q4 bulk + q8 sensitive)
Mixed-precision MLX checkpoint of Kyutai Moshika. Targets ~5 GB of weights, making the 7B Moshi practical on 8 GB Apple Silicon Macs while keeping the quality-critical layers at q8.
Quantization policy
| Path pattern | Bits | Group |
|---|---|---|
text_emb | 8 | 64 |
text_linear | 8 | 64 |
audio_embs.* | 8 | 64 |
depformer.* (all Linear / Embedding) | 8 | 64 |
All transformer.layers.* Linear / Embedding | 4 | 32 |
| Norms, Mimi conv layers | — | — |
The Mimi codec is not part of this checkpoint. Pair with the Mimi tokenizer from kyutai/moshika-mlx-bf16 (tokenizer-e351c8d8-checkpoint125.safetensors).
Memory (weights only)
| Variant | Approx weights | Notes |
|---|---|---|
| BF16 | ~14 GB | Reference |
| q8 (group 64), uniform | ~7.4 GB | kyutai/moshika-mlx-q8 |
| mixed (this) | ~5.0 GB | q4 bulk + q8 sensitive |
| q4 (group 32), uniform | ~3.7 GB | kyutai/moshika-mlx-q4 |
Usage
moshi-swift GUI
Select Moshi q4/q8 mixed in the app's model picker. First run downloads to the HF cache.
moshi-swift CLI
moshi-cli run hf://strumecki/moshika-mlx-mp/model.mp.safetensors \
--config moshi7b \
--mimi-model hf://kyutai/moshika-mlx-bf16/tokenizer-e351c8d8-checkpoint125.safetensors \
--input mic
Reproducibility
Produced from kyutai/moshika-mlx-bf16/model.safetensors using scripts/convert_mixed_precision.py in the moshi-swift fork. Two passes of mlx.nn.quantize with mutually-exclusive class_predicate filters.
Limitations
- All 32 main-LM transformer layers are q4. Quality may differ from uniform q8 — subjective listening tests are recommended before relying on this for production.
- Quantization is structural: the on-disk format records no bits/group_size metadata. Loading in any runtime other than
moshi-swiftrequires applying the same predicate beforeupdate(parameters:). - macOS / Apple Silicon only (mlx-swift requirement).
Attribution
All credit for the underlying model architecture, training, and the Mimi codec to Kyutai and the Moshi paper.
- Library
- mlx
- License
- cc-by-4.0
- Languages
- en
- Created
- May 16, 2026
- Updated
- May 16, 2026