0.23.1

What's Changed

• Fix undefined error id screen upon losing connection to API by @deep1401 in #772
• Fix Output Terminal going blank when re-connecting to another API server by @deep1401 in #775
• Model groups search bar now looks at more things than just model groups name by @deep1401 in #770
• Add no model is running pop-up upon model disconnection to audio screen by @deep1401 in #778
• This fix removes the duplication of native + wavesurfer audio players… by @aliasaria in #776
• Add Top P field for audio by @deep1401 in #777
• Fix/audio cloning upload by @mina-parham in #781
• Dont show nan in text if another model is deleted while downloading by @deep1401 in #784

Full Changelog: v0.23.0...v0.23.1

0.23.0

What's Changed

• Fix native element in audio player by @deep1401 in #748
• Fix no training plugins message and some typos by @deep1401 in #757
• Add/audio cloning by @mina-parham in #746
• Add/adaptor name audio by @mina-parham in #761
• Add text-to-audio as an Audio pipelineTag by @dadmobile in #765
• Add adaptor name in audioplayer by @mina-parham in #762
• Fix/disable audio cloning mlx by @mina-parham in #766
• Add/voice audio by @mina-parham in #749
• Fix Tool Calling by @aahaanmaini in #736

Full Changelog: v0.22.0...v0.23.0

0.22.0

Highlights:

• Alpha release of Text-to-Speech support including generation on CUDA and MLX, and training on CUDA!
• Ability to trigger a workflow through transformerlab-client
• Fix for the stalling download error some users were getting on gated models
• Diffusion has been converted to run as a plugin like other loaders
• Upgrades to key libraries to support newest models (torch, transformers)
• API reorganization to move jobs under experiments

What's Changed

• Remove audio URL from UI after speech is generated by @mina-parham in #720
• Move Jobs Router under Experiment by @aahaanmaini in #701
• Fix Audio Player for datasets preview by @deep1401 in #729
• Fix stopping of jobs by @deep1401 in #739
• Update Old Job Endpoints under Experiment by @aahaanmaini in #740
• Fix/audio engine filtering by @mina-parham in #733
• Convert Image Diffusion to a Plugin by @Sourenm in #628

Full Changelog: v0.21.1...v0.22.0

0.21.1

What's Changed

• Add checkpointing UI by @aliasaria in #711
• Add support for downloading audio datasets from HF by @deep1401 in #709
• Add custom url for tensorboard modal by @deep1401 in #685
• Add/audio tts by @mina-parham in #710
• Change audio output path by @mina-parham in #715

Full Changelog: v0.20.7...v0.21.1

0.20.7

What's Changed

• Improve README.md to show that we support stable diffusion in addition to LLMs
• Fix Plugin Architecture decision in Export Modal
• Improve rendering of job progress
• Add IDE files to gitignore

New Contributors:
@michael-haroon

Full Changelog: v0.20.6...v0.20.7

0.20.5

Potential fix for Windows AutoUpdate

Full Changelog: v0.20.4...v0.20.5

0.20.4

What's Changed

• fix form validation ux for LoRA modal by @MashDevel in #673
• Hide incompatible plugins in dropdown by @MashDevel in #682
• Fix system message by adding a override button by @deep1401 in #670
• Fix CSS for long experiment names by @deep1401 in #688
• Dont install a plugin if no experiment is selected by @deep1401 in #686
• Make plugin install non experiment dependent by @deep1401 in #691
• Show Error When Chat Column Is Not in Valid Message Format by @mina-parham in #695

New Contributors

• @MashDevel made their first contribution in #673

Full Changelog: v0.20.3...v0.20.4

0.20.3

What's Changed

• Don't show CPU recipes on CUDA or MLX by @deep1401 in #679
• version 2 of the list recipes screen by @aliasaria in #677

Full Changelog: v0.20.2...v0.20.3

0.20.2

This is a patch release with some helpful features and bug fixes including:

• training with chat templates: automatically apply chat templates when training on datasets of conversations
• VLM training: use combined image and text datasets on LoRA finetunes
• A number of UI improvements to Training, Generate and Evaluation modals
• fix for docker image so llama.cpp plugin works properly
• plus many more stability and UI fixes

Detailed Log

• Make export page look exactly like evals and generate by @deep1401 in #649
• Fix Close button on Workflow Runs Output Modal by @aahaanmaini in #636
• Fixed Workflow runs Training Jobs Streaming by @aahaanmaini in #634
• Add chat column into config by @mina-parham in #643
• Fix ModelProviderWidget to have an option to also specify loader plugin by @deep1401 in #652
• Fix editing task names in Train and Evaluate by @deep1401 in #647
• Include evaluation steps in GEval Widget by @deep1401 in #658
• Add Progress Bar to Workflow Runs by @aahaanmaini in #656
• Make Workflow JSON Editable in UI by @aahaanmaini in #666
• Use the job streaming route for train outputs too by @deep1401 in #668
• Fix recipe install logic to improve performance for plugins and dataset install by @deep1401 in #665
• Fix selection of tasks with duplicate names and set default mode as TRAIN by @deep1401 in #663
• Fix system message reset by @mina-parham in #638
• Fix blank system message when starting a model by @deep1401 in #669
• Add/activate-recipes-for-everyone by @aliasaria in #601
• Remove Training Recipes from the App by @aliasaria in #660
• Fix Workflow JSON in dark mode by @aahaanmaini in #671
• Workflows can only add tasks within experiment by @aahaanmaini in #648
• Show Apply Chat Template toggle only for specific plugin types by @mina-parham in #657
• Fix Editing, Saving on TRAIN, EVAL and GENERATE by @deep1401 in #674

Full Changelog: v0.20.1...v0.20.2

0.20.1

What's Changed

• Move experimentInfo into a useContext thing by @aliasaria in #630
• Make exporter jobs execute as tasks by @deep1401 in #637
• Add login flow by @dadmobile in #530
• Fix going to step 2 for recipes by @deep1401 in #641
• Send adaptor names when running a model with adaptor by @deep1401 in #644
• Change GetFullPath to getAPIFullPath throughout the codebase by @Sourenm in #646
• Display sweep config by @mina-parham in #605

Full Changelog: v0.20.0...v0.20.1