Results for “nemo-retriever”

Plan, debug, tune, evaluate, export, or deploy public Nemotron embedding and reranking retrieval recipes using the current checkout.

NVIDIA's runtime safety framework for LLM applications. Features jailbreak detection, input/output validation, fact-checking, hallucination detection, PII filtering, toxicity detection. Uses Colang 2.0 DSL for programmable rails. Production-ready, runs on T4 GPU.

Configure fault tolerance, straggler detection, preemption, in-process restart, and re-run state machine for Megatron Bridge training jobs.

Run evaluation tasks against a NeMo Platform server using the Evaluator plugin CLI and Python SDK.

Provides storage and environment conventions for NeMo-RL agents on Brev instances, ensuring large experiment outputs go to /ephemeral and secrets are loaded from .env.

Launch, monitor, stop, and debug NeMo-RL recipes on a Kubernetes cluster using the nrl-k8s CLI, supporting ephemeral and long-lived RayCluster modes.

GPU-accelerated data curation for LLM training, supporting text, image, video, and audio with fuzzy deduplication, quality filtering, semantic deduplication, PII redaction, and NSFW detection.

GPU-accelerated data curation for LLM training. Supports text/image/video/audio. Features fuzzy deduplication (16× faster), quality filtering (30+ heuristics), semantic deduplication, PII redaction, NSFW detection. Scales across GPUs with RAPIDS. Use for preparing high-quality training datasets, cleaning web data, or deduplicating large corpora.

NVIDIA's runtime safety framework for LLM applications. Features jailbreak detection, input/output validation, fact-checking, hallucination detection, PII filtering, toxicity detection. Uses Colang 2.0 DSL for programmable rails. Production-ready, runs on T4 GPU.

GPU-accelerated data curation for LLM training. Supports text/image/video/audio. Features fuzzy deduplication (16× faster), quality filtering (30+ heuristics), semantic deduplication, PII redaction, NSFW detection. Scales across GPUs with RAPIDS. Use for preparing high-quality training datasets, cleaning web data, or deduplicating large corpora.

Guides agents through the full lifecycle of NeMo-RL experiments: understanding recipes, launching reproducible runs, analyzing results, and preserving human oversight with git and TSV logs.

Routes NVIDIA Nemotron Speech (Riva) NIM tasks for ASR, TTS, and NMT, covering cloud-hosted inference, self-hosted Docker deployment, and custom model builds.

Generates custom safety policies for NVIDIA Nemotron content-safety guardrails, producing a Markdown policy, JSON taxonomy, and inference prompts from rough user input.

Run Megatron-LM (MLM) and Megatron Bridge training with mock or real data, covering correlation testing, available recipes, and multi-GPU examples.

Select, validate, patch, and deploy existing NVIDIA Dynamo Kubernetes recipes for model serving with GPU support.

Create and modify NeMo AutoModel training and evaluation recipes, including YAML structure, builders, and execution flow.

Guides AI coding assistants to the official NemoClaw documentation via MCP server or Markdown files for installation, configuration, operation, and troubleshooting.

Fine-tune, evaluate, run inference, and export Cosmos-Embed1 video-text embedding models for tasks like text-to-video retrieval and semantic deduplication.

Train, evaluate, export, prune, and run inference for PointPillars 3D object detection models from LiDAR point clouds using NVIDIA TAO.

Maintain durable session memory across agent disconnects by writing structured checkpoints to the repo's session directory, enabling context recovery.

Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace. Use when training large-scale models with limited compute (5× cost reduction vs dense models), implementing sparse architectures like Mixtral 8x7B or DeepSeek-V3, or scaling model capacity without proportional compute increase. Covers MoE architectures, routing mechanisms, load balancing, expert parallelism, and inference optimization.

Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace. Use when training large-scale models with limited compute (5× cost reduction vs dense models), implementing sparse architectures like Mixtral 8x7B or DeepSeek-V3, or scaling model capacity without proportional compute increase. Covers MoE architectures, routing mechanisms, load balancing, expert parallelism, and inference optimization.

Evaluates LLMs across 100+ benchmarks from 18+ harnesses (MMLU, HumanEval, GSM8K, safety, VLM) with multi-backend execution on local Docker, Slurm HPC, or cloud platforms.

Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace, covering architectures, routing, load balancing, and expert parallelism.

Add programmable safety guardrails to LLM applications at runtime, including jailbreak detection, input/output validation, fact-checking, hallucination detection, PII filtering, and toxicity detection.

Provides guidance for training Mixture-of-Experts models with long context windows, covering context parallelism sizing, selective recomputation, dispatcher choices, and practical patterns from recent experiments.

Train, evaluate, export, and run inference for Masked Auto-Encoder (MAE) models for self-supervised pretraining and fine-tuning of visual representations.

Convert single-node PyTorch distributed scripts into multi-node Slurm sbatch jobs and debug common multi-node failures, covering srun-native and torch.distributed approaches, container setup, NCCL timeouts, and interactive allocation.

This skill should be used when working with reinforcement learning tasks including high-performance RL training, custom environment development, vectorized parallel simulation, multi-agent systems, or integration with existing RL environments (Gymnasium, PettingZoo, Atari, Procgen, etc.). Use this skill for implementing PPO training, creating PufferEnv environments, optimizing RL performance, or developing policies with CNNs/LSTMs.