Skip to content
OpenSmartRoute

Marketplace

Everything your AI needs, in one place.

Ready-made agents, skills, personas, prompts, templates and tools. Each one is checked before it goes live, works with any model, and installs in a click. Rate what you use so the best rises to the top.

143.8K
listings
1
installs
0
reviews
38.7K
publishers
38 results
Skill

flow-nexus-neural

Train and deploy neural networks in distributed E2B sandboxes with Flow Nexus

by ruvnetskills.sh
(0)
0Free
Skill

huggingface-accelerate

Simplest distributed training API. 4 lines to add distributed support to any PyTorch script. Unified API for DeepSpeed/FSDP/Megatron/DDP. Automatic device placement, mixed precision (FP16/BF16/FP8). I

by davila7skills.sh
(0)
0Free
Skill

deepspeed

Expert guidance for distributed training with DeepSpeed - ZeRO optimization stages, pipeline parallelism, FP16/BF16/FP8, 1-bit Adam, sparse attention

by davila7skills.sh
(0)
0Free
Skill

training-llms-megatron

Trains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies. Use when training models >1B parameters, need maximum GPU efficiency (47% MFU on H100

by davila7skills.sh
(0)
0Free
Skill

pytorch-fsdp

Expert guidance for Fully Sharded Data Parallel training with PyTorch FSDP - parameter sharding, mixed precision, CPU offloading, FSDP2

by davila7skills.sh
(0)
0Free
Skill

pytorch-lightning

High-level PyTorch framework with Trainer class, automatic distributed training (DDP/FSDP/DeepSpeed), callbacks system, and minimal boilerplate. Scales from laptop to supercomputer with same code. Use

by davila7skills.sh
(0)
0Free
Skill

ray-train

Distributed training orchestration across clusters. Scales PyTorch/TensorFlow/HuggingFace from laptop to 1000s of nodes. Built-in hyperparameter tuning with Ray Tune, fault tolerance, elastic scaling.

by davila7skills.sh
(0)
0Free
Skill

distributed-llm-pretraining-torchtitan

Provides PyTorch-native distributed LLM pretraining using torchtitan with 4D parallelism (FSDP2, TP, PP, CP). Use when pretraining Llama 3.1, DeepSeek V3, or custom models at scale from 8 to 512+ GPUs

by davila7skills.sh
(0)
0Free
Skill

verl-rl-training

Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL). Use when implementing RLHF, GRPO, PPO, or other RL algorithms for LLM post-training at scale with flexib

by davila7skills.sh
(0)
0Free
Skill

nemo-automodel-distributed-training

Guide for selecting and configuring distributed training strategies in NeMo AutoModel, including FSDP2, Megatron FSDP, DDP, and parallelism settings.

by nvidiaskills.sh
(0)
0Free
Skill

physicsnemo-shard-tensor

Official NVIDIA-authored guidance for PhysicsNeMo ShardTensor domain parallelism — integrate domain parallelism into training/inference scripts (new or existing) with DDP or FSDP2, write and register

by nvidiaskills.sh
(0)
0Free
Skill

distributed-llm-pretraining-torchtitan

Provides PyTorch-native distributed LLM pretraining using torchtitan with 4D parallelism (FSDP2, TP, PP, CP). Use when pretraining Llama 3.1, DeepSeek V3, or custom models at scale from 8 to 512+ GPUs

by orchestra-researchskills.sh
(0)
0Free
Skill

verl-rl-training

Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL). Use when implementing RLHF, GRPO, PPO, or other RL algorithms for LLM post-training at scale with flexib

by orchestra-researchskills.sh
(0)
0Free
Skill

huggingface-accelerate

Simplest distributed training API. 4 lines to add distributed support to any PyTorch script. Unified API for DeepSpeed/FSDP/Megatron/DDP. Automatic device placement, mixed precision (FP16/BF16/FP8). I

by orchestra-researchskills.sh
(0)
0Free
Skill

deepspeed

Expert guidance for distributed training with DeepSpeed - ZeRO optimization stages, pipeline parallelism, FP16/BF16/FP8, 1-bit Adam, sparse attention

by orchestra-researchskills.sh
(0)
0Free
Skill

training-llms-megatron

Trains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies. Use when training models >1B parameters, need maximum GPU efficiency (47% MFU on H100

by orchestra-researchskills.sh
(0)
0Free
Skill

pytorch-fsdp2

Adds PyTorch FSDP2 (fully_shard) to training scripts with correct init, sharding, mixed precision/offload config, and distributed checkpointing. Use when models exceed single-GPU memory or when you ne

by orchestra-researchskills.sh
(0)
0Free
Skill

pytorch-lightning

High-level PyTorch framework with Trainer class, automatic distributed training (DDP/FSDP/DeepSpeed), callbacks system, and minimal boilerplate. Scales from laptop to supercomputer with same code. Use

by orchestra-researchskills.sh
(0)
0Free
Skill

ray-train

Distributed training orchestration across clusters. Scales PyTorch/TensorFlow/HuggingFace from laptop to 1000s of nodes. Built-in hyperparameter tuning with Ray Tune, fault tolerance, elastic scaling.

by orchestra-researchskills.sh
(0)
0Free
Skill

flow-nexus-neural

Train and deploy neural networks in distributed E2B sandboxes with Flow Nexus

by proffesor-for-testingskills.sh
(0)
0Free
Skill

huggingface-accelerate

Simplest distributed training API. 4 lines to add distributed support to any PyTorch script. Unified API for DeepSpeed/FSDP/Megatron/DDP. Automatic device placement, mixed precision (FP16/BF16/FP8). I

by ovachieverskills.sh
(0)
0Free
Skill

deepspeed

Expert guidance for distributed training with DeepSpeed - ZeRO optimization stages, pipeline parallelism, FP16/BF16/FP8, 1-bit Adam, sparse attention

by ovachieverskills.sh
(0)
0Free
Skill

training-llms-megatron

Trains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies. Use when training models >1B parameters, need maximum GPU efficiency (47% MFU on H100

by ovachieverskills.sh
(0)
0Free
Skill

pytorch-fsdp

Expert guidance for Fully Sharded Data Parallel training with PyTorch FSDP - parameter sharding, mixed precision, CPU offloading, FSDP2

by ovachieverskills.sh
(0)
0Free
1

Find

Search or browse by kind. Every card shows who made it, how many people installed it and what they think.

2

Install

One click. You get a manifest the router understands, plus copy-paste snippets for the CLI, Python and YAML.

3

Rate and publish

Leave a star rating after you have used it. Made something useful? Publish it - free listings go live immediately.

Prefer the terminal? osr stack apply registry://starter installs the starter template.