
Guides an agent through enabling Megatron FSDP sharded training in Megatron-Bridge
- Open source
- 4K GitHub stars
- AI models and LLM tooling

Guides an agent through enabling Megatron FSDP sharded training in Megatron-Bridge

Guides an agent through reducing peak GPU memory in Megatron Bridge training runs

Guides tuning of MoE expert-parallel communication overlap in Megatron Bridge

Guides choice of MoE token dispatcher (alltoall, DeepEP, HybridEP) by hardware and EP size

Hardware-specific MoE training playbooks and candidate parallelism configs for Megatron

Guides long-context MoE training in Megatron Bridge: CP sizing, recompute and dispatcher

Guidance for choosing FSDP or 3D parallelism when training MoE vision-language models

Guides choosing and combining TP, DP, PP, CP and EP degrees for Megatron Bridge training

Guides Megatron-Bridge sequence packing and long-context training setup and validation

Guides enabling and verifying TP, DP and PP communication overlap in Megatron-Bridge

Picks Megatron Bridge training recipes and resizes parallelism for your GPU allocation

Runs iterative NeMo-RL experiments with git branches and TSV logs as a research ledger

Guides an agent through improving NVIDIA Riva/NeMo speech recognition for a domain

Plans and configures chained Nemotron customization steps like SFT, RL, and evaluation

Guides an agent through NVIDIA Nemotron embed and rerank retrieval recipes safely

Guides an agent through deploying and testing NVIDIA Nemotron Speech ASR, TTS and NMT

Builds and fixes NVIDIA speech-to-speech voice agents on Pipecat or LiveKit

Finetunes NV-Segment-CT VISTA3D on your CT NIfTI image and label datasets

Runs a guarded agent loop that tunes an existing NVFLARE federated job for a metric

Builds a final report and JSON summary for a stopped NVFLARE Auto-FL campaign

Converts Hugging Face Trainer or TRL SFTTrainer code into an NVFLARE federated job

Converts existing PyTorch Lightning training code into a federated NVFLARE job

Converts plain PyTorch training code into a federated NVFLARE job

Shared NVFLARE reference files and templates that other NVFLARE skills load on demand

Runs NVIDIA PAIDF Cosmos Predict to generate videos from a media sample list

Benchmarks a RAG pipeline's answer quality on local corpus and train.json datasets

Finds false-positive and false-negative yes/no VLM predictions for root-cause analysis

Prepares and validates a dataset and recipe for fine-tuning AnomalyGenNext LoRA weights

Fine-tunes and runs NVIDIA CLIP models for image-text retrieval and zero-shot

Runs Cosmos-Embed1 video-text embedding for search, dedup and fine-tuning in Docker
See every kind of AI for ai models and llm tooling: tools, agents, MCP servers and skills.Ranked: the best AI for ai models and llm tooling.