// Use when the user wants to estimate GPU memory (VRAM) requirements for a training configuration, check if a model will fit on their GPUs, or plan GPU allocation for training.
| name | memory-estimation |
| description | Use when the user wants to estimate GPU memory (VRAM) requirements for a training configuration, check if a model will fit on their GPUs, or plan GPU allocation for training. |
| allowed-tools | ["Bash(${CLAUDE_PLUGIN_ROOT}/scripts/th_estimate.sh:*)","Bash(${CLAUDE_PLUGIN_ROOT}/scripts/th_detect.sh:*)"] |
Estimate GPU VRAM requirements before committing to a training run.
"${CLAUDE_PLUGIN_ROOT}/scripts/th_detect.sh"
If library=missing, tell the user to install training_hub first via the setup-guide skill.
Execute the estimation script with user-provided parameters or config defaults:
"${CLAUDE_PLUGIN_ROOT}/scripts/th_estimate.sh" $ARGUMENTS
Parse the JSON output and present clearly:
max_seq_len (e.g., 4096 -> 2048)effective_batch_size| Method | For | Estimator |
|---|---|---|
basic | SFT, GRPO | BasicEstimator |
osft | OSFT | OSFTEstimator |
lora | LoRA-SFT, LoRA-GRPO | LoRAEstimator |
qlora | Quantized LoRA | QLoRAEstimator |
If no method is specified, the script infers it from the configured algorithm.
Use when the user wants to set up LLM training for the first time, or when training_hub is not yet installed/configured in the current environment.
Use when the user wants to run a training job using a saved configuration. For algorithm selection, hyperparameter advice, or troubleshooting, use the training-hub-guide skill instead.
Guides users through LLM post-training with Training Hub, including installation, algorithm selection (SFT, OSFT, LoRA), hyperparameter tuning, troubleshooting OOM errors, interpreting loss curves, and leveraging backend-specific features. Use when the user is working with training_hub, fine-tuning language models, asking about SFT/OSFT/LoRA training, or debugging GPU/CUDA training issues.