The doctor is in
Gheorghe Chesler
nightmedia
AI & ML interests
Quanting destroys quality. What if it improves it?
I am exploring creative ways to reach a size/quality compromise that also improve the average performance byte per byte. Focusing the model.
Recent Activity
updated
a model
about 1 hour ago
nightmedia/Jan-nano-dwq6-mlx
updated
a model
about 2 hours ago
nightmedia/Jan-nano-dwq5-mlx
published
a model
about 5 hours ago
nightmedia/Mistral-Devstral-2505-CODER-Brainstorm40x-44B-q4-mlx
Organizations
None yet
models
130
nightmedia/Jan-nano-dwq6-mlx
Text Generation
•
1B
•
Updated
nightmedia/Jan-nano-dwq5-mlx
Text Generation
•
0.9B
•
Updated
nightmedia/Mistral-Devstral-2505-CODER-Brainstorm40x-44B-q4-mlx
Updated
nightmedia/Mistral-Devstral-2505-CODER-Brainstorm40x-44B-q5-mlx
Updated
nightmedia/Mistral-Devstral-2505-CODER-Brainstorm40x-44B-q8-mlx
Updated
nightmedia/Jan-nano-dq58-mlx
Text Generation
•
0.9B
•
Updated
nightmedia/Jan-nano-dq48-mlx
Text Generation
•
0.8B
•
Updated
nightmedia/Jan-nano-dwq4-mlx
Text Generation
•
0.8B
•
Updated
nightmedia/Mistral-Devstral-2505-CODER-Brainstorm40x-44B-q6-mlx
Text Generation
•
44B
•
Updated
nightmedia/Jan-nano-dq38-mlx
Text Generation
•
0.8B
•
Updated
datasets
0
None public yet