27 lines
496 B
Markdown
27 lines
496 B
Markdown
|
# Narrative Memory Filter
|
||
|
|
|
||
|
|
**0.0.2:**
|
||
|
|
- More reliable way of getting chat IDs.
|
||
|
|
|
||
|
|
**0.0.1:**
|
||
|
|
- Initial proof-of-concept release.
|
||
|
|
|
||
|
|
|
||
|
|
# GPU Scaling Filter
|
||
|
|
|
||
|
|
**0.2.0:**
|
||
|
|
- Fix filter not working when using base models directly.
|
||
|
|
- Do not fire the filter if model is not owned by ollama.
|
||
|
|
|
||
|
|
**0.1.0:**
|
||
|
|
- Initial release.
|
||
|
|
|
||
|
|
# Output Sanitization Filter
|
||
|
|
|
||
|
|
**0.2.0:**
|
||
|
|
- Prevent errors when an LLM reply is updated by the filter (i.e.
|
||
|
|
return body from the outlet function).
|
||
|
|
|
||
|
|
**0.1.0:**
|
||
|
|
- Initial Release
|