Update README.md

remove debugging

format prompts and add debug log

make hub download once per app load

support loading leaderboard/votes from persistent source

standardize new scheduled locations

shorten scheduler commit window for testing

refresh and persistent logging

update data pipeline

remove cogito for gpu-hour reasons and add approved llama models

add additional models

fix typo

gemma 3 4b misaligned bug, so removing that as an option for now

update model list

revert to sequential processing

typo fix

prepare for zeroGPU

specified device

vllm backend swap v1

Add manual clean up of the resources

Removed JS (#20)

Delete static/js (#21)

updated interruption handling for the "try a new question" + user experience improvement (#16)

UI fixes (#17)

add js file to handle scrolling (#18)

removed backup vote_details (#19)

Refine UI (#14)

Refined writing and layout (#15)

proper threaded generation interrupt

Rewrote interruption & improved UI (#9)

Handled interruption (#10)

Improved UI (#11)

Added shared interruption flag (#12)

ready hyperlinks for leaderboard

Remove timeout & update elo ranking (#4)

Rewrote the processor to make it more robust to edge cases (#5)

Implemented elo (#6)

Minor UI fixes and styling for ranking (#7)

Component to store vote details to a csv (#8)

sanitized arena_df

first pass at async handling

switched models to using full context

Added Context preload and inference interruption (#2)

Support nigh mode and mobile view (#3)

update data loading (#1)

fix models

remove gated models for now

remove fa2 req

v1 inference code

Remove .DS_Store files and add gitignore