Downloads · 30 days
44
5% of all-time downloads
GigaChat3.1-10B-A1.8B-MLX-oQ8
deepsweet/GigaChat3.1-10B-A1.8B-MLX-oQ8
GigaChat3.1-10B-A1.8B-MLX-oQ8 is a text generation model from deepsweet. Use it when you need the model to write or continue text. It is set up for mlx. The card lists the license as mit.
This model was converted to MLX format from ai-sage/GigaChat3.1-10B-A1.8B using oMLX v0.2.24 with oQ Quantization.
All-time downloads
874
Public
Parameters
10.7B
11 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors11 GB · 100%
Parameter types
How the weights are stored.
U3210.7B · 100%
Base models
Model card
From the Hugging Face model README
This model was converted to MLX format from ai-sage/GigaChat3.1-10B-A1.8B using oMLX v0.2.24 with oQ Quantization.
[!WARNING] Multi-Token Prediction (MTP) had to be disabled (
"num_nextn_predict_layers": 0) and related layers had to be removed (model.layers.26.*).
Thanks RockTalk/GigaChat3.1-10B-A1.8B-MLX-4bit for the tip.
Settings:
- Level: oQ8
- Sensitivity model: none
- Text Only: yes
- Task
- Text Generation
- Library
- mlx
- Type
- deepseek_v3
- License
- mit
- Languages
- ru, en
- Created
- Mar 28, 2026
- Updated
- Apr 9, 2026