Commit History

No startup
e8ba1ec
Running

Tim Luka Horstmann commited on

No think
8356d0c

Tim Luka Horstmann commited on

Updated cache dir
220ed89

Tim Luka Horstmann commited on

Updated install
63e32ac

Tim Luka Horstmann commited on

Try qwen3
e47a0a3

Tim Luka Horstmann commited on

Update stay alive
703cd97

Tim Luka Horstmann commited on

Faster model
a79e01b

Tim Luka Horstmann commited on

Smaller 3B model
09c93a8

Tim Luka Horstmann commited on

use mlock
f2054a9

Tim Luka Horstmann commited on

Add ram usage endpoint
687de1a

Tim Luka Horstmann commited on

Fix contxt
e112ae1

Tim Luka Horstmann commited on

Fixed
3bbf0cd

Tim Luka Horstmann commited on

Updated to use history
6f6e59d

Tim Luka Horstmann commited on

increased batch size again
58d2235

Tim Luka Horstmann commited on

Fixed path
dc475e9

Tim Luka Horstmann commited on

Bigger model
54039cd

Tim Luka Horstmann commited on

added date
b77d28c

Tim Luka Horstmann commited on

Increased context window
051899e

Tim Luka Horstmann commited on

Added CV
c020f09

Tim Luka Horstmann commited on

fix link
b6763a9

Tim Luka Horstmann commited on

No RAG
95c3613

Tim Luka Horstmann commited on

Updated straeming
392cd96

Tim Luka Horstmann commited on

Better streaming and less hallucinations.
8583b57

Tim Luka Horstmann commited on

deploy
655702e

Tim Luka Horstmann commited on

restart
c861c71

Tim Luka Horstmann commited on

fixed streaming
59b5835

Tim Luka Horstmann commited on

trigger rebuild
bdbefdd

Tim Luka Horstmann commited on

Updated backend with chat completion
293413b

Tim Luka Horstmann commited on

Cogito model
48a65b5

Tim Luka Horstmann commited on

Updated to avoid weird outputs
e54e8f7

Tim Luka Horstmann commited on

Improved model and RAG
9c89db3

Tim Luka Horstmann commited on

Switched to Llama-3.2-1B Q4_K, added impersonation, optimized performance
a29c4ff

Tim Luka Horstmann commited on

Switch to llama
83ec808

Tim Luka Horstmann commited on

Speedup
a2d5223

Tim Luka Horstmann commited on

bigger model
bd95004

Tim Luka Horstmann commited on

Fixed Qwen 2.5 1.5B with llama_cpp for HF Spaces
58272f8

Tim Luka Horstmann commited on

updated to llama
aa6b888

Tim Luka Horstmann commited on

updated app to use ctransformers and gemma
b845672

Tim Luka Horstmann commited on

updated app to use ctransformers and gemma
587f13e

Tim Luka Horstmann commited on

updated app to use ctransformers and gemma
1faeb4c

Tim Luka Horstmann commited on

Updated app to use ctransformers and gemma without token
588cb6a

Tim Luka Horstmann commited on

updated dockerfile
a7ba255

Tim Luka Horstmann commited on

Initial setup
cb8303f

Tim Luka Horstmann commited on

initial commit
61a9825
verified

Luka512 commited on