Hugging Face
Models
Datasets
Spaces
Buckets
Docs
Enterprise
Pricing
Website
Tasks
HuggingChat
Collections
Languages
Organizations
Community
Blog
Posts
Daily Papers
Hardware
Learn
Discord
Forum
GitHub
Solutions
Team & Enterprise
Hugging Face PRO
Enterprise Support
Inference Providers
Inference Endpoints
Storage Buckets
Log In
Sign Up
Spaces:
Cnass
/
mnaga
Like
0
Sleeping
App
Files
Files
Community
Fetching metadata from the HF Docker repository...
main
mnaga
Commit History
Sep 24
13:07
Update Dockerfile
Cnass
Verified signature
40a434a
10:35
Update Dockerfile
Cnass
Verified signature
260d67c
Sep 17
14:15
Fix inpainting VRAM OOM: cap resolution, retry ladder, allocator hint
Cnass
Verified signature
4d38a10
13:09
Fix: enum-safe model prefetch (docker_prepare filter is a no-op on py3.11)
Cnass
Verified signature
53f0d6c
13:01
Fix push OOM: single CUDA runtime copy, split large layers
Cnass
Verified signature
70f714f
12:48
Fix: Ubuntu 24.04 base (GLIBC 2.38 for llama.cpp CUDA) + Python 3.11
Cnass
Verified signature
923aa02
12:44
Fix: install libgomp1 so llama-server can link
Cnass
Verified signature
afd510d
12:38
Build from github.com/Waifuhtr/mnaga: manga -> Turkish via Hy-MT2-7B
Cnass
Verified signature
35719f8
Minimal Space: Dockerfile + metadata only. Application source is cloned from GitHub at build time; all model files are baked into image layers so switching CPU -> T4 re-downloads nothing.
12:07
initial commit
Cnass
Verified signature
f9fdf8a
