feffy380

write_to_file arguments are printed on a single line, which is hard to read

6

### Duplicates - [X] I have searched the existing issues ### Summary 💡 When the current COMMAND is write_to_file (and possibly others), large multiline arguments should be printed as multiple...

The current implementation only excludes inputs matching the memory key. When using CombinedMemory, there will be multiple keys and the vector store memory will save everything except the memory_key. This...

[Bug]: Query on DuckDB returns less than k results

1

### What happened? I consistently get fewer documents than requested when querying a DuckDB collection. For example, querying for 100 documents in a collection of 300+ only returns about 20-30....

e621: support "lore" tag category

1

**Is your feature request related to a problem? Please describe** e621 has a `lore` tag category with id=8 (see https://e621.net/help/api under Tags > Categories). Grabber doesn't recognize this category, so...

Faster training with token downsampling

4

This PR adds support for training with [token downsampling](https://github.com/ethansmith2000/ImprovedTokenMerge) and replaces my token merge PR (#1146). Token downsampling is a lossy optimization that significantly speeds up inference and training. It...

**The bug** Trying to use a model through llama.cpp built with hipBLAS leads to an immediate crash: ```bash CUDA error: shared object initialization failed current device: 0, in function ggml_cuda_op_mul_mat...

### Checklist - [X] The issue exists after disabling all extensions - [X] The issue exists on a clean installation of webui - [ ] The issue is caused by...