Downloads · 30 days
21
15% of all-time downloads
Gemma-4-E4B-Nightcap-MLX-8bit
andyoneal/Gemma-4-E4B-Nightcap-MLX-8bit
Gemma-4-E4B-Nightcap-MLX-8bit is a text generation model from andyoneal. Use it when you need the model to write or continue text. It is set up for mlx. The card lists the license as gemma.
MLX quantization of andyoneal/Gemma-4-E4B-Nightcap, a roleplay / creative-writing task-arithmetic merge of Gemma 4 E4B. See the main repo for the full card — what's in it, how it behaves, and how the recipe was chosen.
All-time downloads
137
Public
Parameters
7.5B
8 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors7.9 GB · 100%
Parameter types
How the weights are stored.
U327.5B · 100%
Try a prompt
Base models
Model card
From the Hugging Face model README
Gemma 4 E4B Nightcap — MLX 8-bit
MLX quantization of andyoneal/Gemma-4-E4B-Nightcap, a roleplay / creative-writing task-arithmetic merge of Gemma 4 E4B. See the main repo for the full card — what's in it, how it behaves, and how the recipe was chosen.
Text-only (the conversion keeps the language stack); chat template included.
| variant | bits/weight | size | notes |
|---|---|---|---|
| this repo | 8.501 | ~7.4 GB | ~8 GB peak memory, ~50 tok/s on Apple silicon |
uv run --with mlx-lm python -m mlx_lm generate \
--model andyoneal/Gemma-4-E4B-Nightcap-MLX-8bit \
--system-prompt "You are..." --prompt "..." \
--chat-template-config '{"enable_thinking": false}' --temp 1.0
Samplers: temp 1.0, min-p 0.02, no repetition penalties (they do more harm than good on this family). Temp 0.8 / top-p 0.95 also works fine.
- Task
- Text Generation
- Library
- mlx
- Type
- gemma4
- License
- gemma
- Languages
- en
- Created
- Jul 19, 2026
- Updated
- Jul 19, 2026