Model Training skills
Free agent skills tagged model training, ready to install into any SKILL.md-compatible agent.
18 skills
MoE Optimization Workflow
nvidia
Streamline MoE model training with systematic guidance.
Activation Recompute
nvidia
Optimize GPU memory usage with activation recompute.
Parallelism Strategy Selection
nvidia
Optimize parallelism for Megatron Bridge models effectively.
Megatron Bridge Recipe Recommender
nvidia
Streamline your model training with tailored recipe suggestions.
Masked Autoencoder Training
nvidia
Efficiently pretrain and fine-tune visual models.
PyTorch Lightning
k-dense-ai
Streamline your deep learning workflows with PyTorch.
GRPO & RLVR Training
wshobson
Enhance model reasoning with verifiable rewards.
Fine-Tuning Expert
jeffallan
Optimize LLM training with parameter-efficient methods.
Trace To Training Data
wshobson
Transform evaluation traces into training data efficiently.
Hugging Face LLM Trainer
huggingface
Train and fine-tune models on Hugging Face Jobs effortlessly.
Train Sentence Transformers
huggingface
Efficiently train and fine-tune sentence-transformers models.
Uni-Mol Tools
hkuds
Interactive CLI for molecular property prediction workflows.
Megatron FSDP
nvidia
Optimize distributed training with Megatron FSDP.
Simple Preference Optimization
nousresearch
Optimize preferences without reference models.
Cloud Backtest for Neural Trader
ruvnet
Run intensive trading simulations in the cloud.
TRL Training
huggingface
Train and fine-tune transformer models with ease.
Add Training Support
civitai
Integrate ecosystems into LoRA training seamlessly.
Unsloth
nousresearch
Accelerate LoRA/QLoRA fine-tuning with less VRAM.
