Commit History
fix: logging issues because of type conflict that was introduced when adding new scheduler
8799da1
Philip Kehl
commited on
feat: added jseq scheduler to javascript app
c9e2513
Philip Kehl
commited on
fix: added error catch in request manager, adjusted cloud service getModelName method and inserted a console log for the no_think add-on
1d85bd9
Philip Kehl
commited on
fix: fixed qwen3 check to check first if service.getModelName() is defined
fd2aa2a
nicolasfwyss
commited on
Merge branch 'main' of github.com:fabianHueni/UNIBE_MSGAI_HS25_Project
3cce2bf
Philip Kehl
commited on
feat: disable thinking for qwen model
578c052
Philip Kehl
commited on
feat: enhance download of csv after experiment completion
281044f
chore: added comments
752ec01
Philip Kehl
commited on
feat: added stripping method for thinking tokens, increased maxNewTokens to 500, set enableThinking to false in onDeviceService
60c1285
Philip Kehl
commited on
feat: add queues for on device and cloud processing to measure queue time
a2101c1
feat: track queueing time, inference time, and total latency separately. the timestamp for start inference should be taken somewhere else
0314e5b
Philip Kehl
commited on
fix: changed the prompt to include the passage/context as well and moved the full prompt to scheduler.js file
9b3ecab
Nicolas Wyss
commited on
fix: correct shown device stats and add message template to on device service
714c7da
feat: implement CSV download and converted logs to a table
0726d27
feat: fix prediction output object for evaluation
408650b
feat: implement OpenRouter as cloud model provider, optimize UI, fix some issues
7b5fcf7
feat: model download progress bar and added metrics f1WordLevel, Rouge1, TotalTokens, TokensPerSecond
048bfac
Philip Kehl
commited on