h2oai / h2o-llmstudio

H2O LLM Studio - a framework and no-code GUI for fine-tuning LLMs. Documentation: https://docs.h2o.ai/h2o-llmstudio/

h2oai/h2o-llmstudio Issues

[BUG]httpx.ReadTimeout && httpcore.ReadTimeout
Updated 9 days ago1
[UI] Improve plot of validation loss and metric
Updated 10 days ago
[FEATURE] Update to wave 1.4.0 and add wildcard option
Closed 11 days ago
[TEST] Test gated HF repository functionality
Updated 11 days ago
[FEATURE] Option to plot train/eval plots with epoch instead of step on x-axis
Updated 12 days ago
[FEATURE] Multilabel classification
Closed 13 days ago
[BUG] S3 import from public bucket fails on first try
Updated 16 days ago
[BUG] Import HF mentions Kaggle secret key
Closed 17 days ago
[FEATURE] Add an early `nvidia-smi` call at the startup routine to expose GPU issues
Closed 17 days ago
[BUG] Purely cosmetic issue, progress values are not rounded
Updated 18 days ago1
[FEATURE] Prepackage demo datasets in docker image
Updated 19 days ago
[FEATURE] Export Chat session
Closed 23 days ago
[FEATURE] Add Danube3 to the default model list
Closed 23 days ago1
[BUG] Error during LoRA-merge in HF upload for Llama 3.1 70B model
Updated 23 days ago21
[BUG] HF_Transfer is only applied to download of models
Updated 23 days ago
[BUG] Docker image for v1.9.0 doesn't contain pipenv
Closed 23 days ago
[FEATURE] Add Rank Stabilization Scaling Factor for Fine-Tuning with LoRA
Closed 25 days ago
[FEATURE] Support common conversation data format
Updated a month ago
[BUG] Frequent failure to push checkpoint to Hugging Face
Closed a month ago4
[FEATURE] Add (experimental) FP8 support
Updated a month ago
Default datasets for classification and regression
Updated a month ago
[BUG] Memory allocation left resident in GPU(s) after model upload to HuggingFace
Updated a month ago1
[UX] Screen hangs when you click Download Model
Updated a month ago1
Compare Zero-Epoch Prediction with Fine-Tuned Prediction as well as Validation Score Comparison
Closed a month ago1
Request to add setup-dev-no-flash to MakeFile
Closed a month ago1
[DOC] Document Importing datasets with H2O Drive
Updated a month ago
[FEATURE] Support for Llama 3.1
Closed a month ago
[FEATURE] More options for attention implementation
Closed a month ago1
[FEATURE] Importing Data From H2O Drive to LLMStudio
Closed a month ago
[FEATURE] Implement SimPO
Closed a month ago
H2O LLM Studio error "Unknown exception" when running app
Closed a month ago2
Can't see the local saved model in the 'LLM backbone' pull down menu of 'Create Experiment'
Closed a month ago4
[BUG] Backslash of prompt column separator gets applied multiple times when reload is triggered
Closed a month ago
Can't access meta-llama/Llama-2-7b using llmstudio cli tool
Closed a month ago1
[BUG] Training DeepSeek-Coder-V2-Lite-Base MOE inordinately slow
Updated a month ago1
[DOC] Document Adding New Backbone
Closed 2 months ago2
[BUG] GPU id error when restarting on a different instance
Updated 2 months ago
[FEATURE] Add Stopping Criteria to Grid Search
Updated 2 months ago3
[FEATURE] Apple Silicon: Add support of the mlx-lm
Updated 2 months ago1
[BUG] Default model is added twice to the LLM Backbone list
Closed 3 months ago
[BUG] Deepseek tokenizer error on main (worked before)
Closed 3 months ago4
[DOCS] Duplicate Questions in the FAQ's
Closed 3 months ago
[FEATURE] Freezing layers
Closed 3 months ago
[FEATURE] Connection with LLM DataStudio
Closed 3 months ago1
[FEATURE] Select multiple training dataframes
Updated 3 months ago
[BUG] Code rendering in the validation prediction insights replaces characters
Updated 3 months ago
[BUG] HuggingFace export does not preserve bfloat16 weights but converts to float16 silently when using CPU for upload
Closed 3 months ago9
[CODE IMPROVEMENT] Custom HF model for classification
Updated 3 months ago
[FEATURE] add support for multigpu, splitting model across gpus without using deepspeed/fsdp
Closed 3 months ago3
Data Format section has a broken link
Closed 3 months ago4