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

[FEATURE] Support Windows Docker/WSL
Updated 9 days ago1
[UX] Missing ID column when a parent_id was specified, give a nice warning instead of a traceback
Closed a month ago3
[BUG] Loading big dataset with 2 million examples is causing an error
Closed a month ago4
[CODE IMPROVEMENT] Add notice when the number of samples is cut in the Validation Insights tab
Updated a month ago1
[BUG] Unnecessary UI scrollbars
Closed a month ago
[DOC] Add to FAQ, facing OSError: Consistency check failed in experiment
Closed a month ago1
[DOC] Expand dataset import connectors
Closed a month ago
[FEATURE] Option to override the default model lists via ENV var
Closed 2 months ago
[DOC] Document Importing datasets with H2O Drive
Closed 2 months ago
[BUG] HF_Transfer is only applied to download of models
Closed 2 months ago
[UI] Improve plot of validation loss and metric
Updated 2 months ago
[BUG] Experiment Progress shows 100% but status Running with an ETA > 0
Closed 2 months ago2
[BUG] Unable to run H2O LLM Studio CLI in T4 Colab
Closed 2 months ago2
CONTAINER - Enable remote access via environmental variable
Closed 2 months ago1
[FEATURE] Adapt "Chat" tab for classification and regression use-cases
Updated 3 months ago
[BUG]httpx.ReadTimeout && httpcore.ReadTimeout
Closed 3 months ago2
[BUG] Error during LoRA-merge in HF upload for Llama 3.1 70B model
Updated 3 months ago22
[CODE IMPROVEMENT] Add warning if experiment name already exists
Updated 3 months ago
[FEATURE] Prepackage demo datasets in docker image
Closed 3 months ago
[CHORE] Add automated tests for Docker build process and execution
Updated 3 months ago
Default datasets for classification and regression
Closed 3 months ago1
[FEATURE] Update to wave 1.4.0 and add wildcard option
Closed 3 months ago
[TEST] Test gated HF repository functionality
Updated 3 months ago
[FEATURE] Multilabel classification
Closed 3 months ago
[BUG] S3 import from public bucket fails on first try
Updated 3 months ago
[BUG] Import HF mentions Kaggle secret key
Closed 3 months ago
[FEATURE] Add an early `nvidia-smi` call at the startup routine to expose GPU issues
Closed 3 months ago
[BUG] Purely cosmetic issue, progress values are not rounded
Updated 3 months ago1
[FEATURE] Export Chat session
Closed 4 months ago
[FEATURE] Add Danube3 to the default model list
Closed 4 months ago1
[BUG] Docker image for v1.9.0 doesn't contain pipenv
Closed 4 months ago
[FEATURE] Add Rank Stabilization Scaling Factor for Fine-Tuning with LoRA
Closed 4 months ago
[FEATURE] Support common conversation data format
Updated 4 months ago
[BUG] Frequent failure to push checkpoint to Hugging Face
Closed 4 months ago4
[FEATURE] Add (experimental) FP8 support
Updated 4 months ago
[BUG] Memory allocation left resident in GPU(s) after model upload to HuggingFace
Updated 4 months ago1
Request to add setup-dev-no-flash to MakeFile
Closed 4 months ago1
[FEATURE] Support for Llama 3.1
Closed 4 months ago
[FEATURE] More options for attention implementation
Closed 4 months ago1
[FEATURE] Importing Data From H2O Drive to LLMStudio
Closed 4 months ago
H2O LLM Studio error "Unknown exception" when running app
Closed 4 months ago2
Can't see the local saved model in the 'LLM backbone' pull down menu of 'Create Experiment'
Closed 4 months ago4
[BUG] Backslash of prompt column separator gets applied multiple times when reload is triggered
Closed 4 months ago
[BUG] Training DeepSeek-Coder-V2-Lite-Base MOE inordinately slow
Updated 4 months ago1
[DOC] Document Adding New Backbone
Closed 5 months ago2
[BUG] GPU id error when restarting on a different instance
Updated 5 months ago
[FEATURE] Add Stopping Criteria to Grid Search
Updated 5 months ago3
[FEATURE] Apple Silicon: Add support of the mlx-lm
Updated 5 months ago1
[BUG] Default model is added twice to the LLM Backbone list
Closed 5 months ago
[BUG] Deepseek tokenizer error on main (worked before)
Closed 6 months ago4