TRL: SFT, DPO, GRPO, RLOO reward modeling for LLM RLHF.

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library in…

Guide on efficiently fine-tuning large language models using LoRA adapters with Python code examples and configuration details.

Fine-tune LLMs with data preparation, provider selection, cost estimation, evaluation, and compliance checks.

Indexed by skills.sh from jeffallan/claude-skills

Indexed by skills.sh from orchestra-research/ai-research-skills

Indexed by skills.sh from orchestra-research/ai-research-skills

Indexed by skills.sh from davila7/claude-code-templates

Indexed by skills.sh from davila7/claude-code-templates

Cosmos-Embed1 video-text embedding for text-to-video retrieval, video-to-video search, semantic deduplication, and fine-tuning. Use when the user asks to "fine-tune Cosmos-Embed1", "run cosmos-embed inference", "export Cosmos-Embed1", "embed videos", or "search videos with text".

Axolotl: YAML LLM fine-tuning (LoRA, DPO, GRPO).

Unsloth: 2-5x faster LoRA/QLoRA fine-tuning, less VRAM.

AI tuning assistant for Gran Turismo 7. Send a photo of your GT7 settings screen → get a complete optimized tune via GT Pro Tune calculator. Supports all 581...

Deep performance tuning workflow—goals and measurement, profiling, hotspots, caching and concurrency trade-offs, system-specific tuning (DB, GC, network), an...

Calculate PI/PID controller gains using Internal Model Control (IMC) tuning rules for first-order systems.

Use when planning, debugging, tuning, evaluating, exporting, or deploying public Nemotron `embed`/`rerank` retrieval recipes.

模型微调服务 | Model Fine-tuning Service. LLM LoRA/QLoRA 微调 | LLM LoRA/QLoRA fine-tuning. 7B/13B 模型微调 | 7B/13B model fine-tuning. Stable Diffusion LoRA 训练 | Stable...

A battle-tested OpenClaw setup with pre-wired identity, memory architecture, security protocols, and automation. Skip weeks of trial and error — install this...

Collaboratively design, evaluate, iterate on, and recommend a final launch candidate for a target prompt under the principle of “human-gated, agent-executed”...

Adjust tradebot parameters incrementally to improve tradability safely based on market regime and diagnostic signals while maintaining risk controls.

Techniques for reducing peak GPU memory in Megatron Bridge — expandable segments, PEFT + SP input re-gather, parallelism resizing, activation recompute, CPU offloading constraints, and common OOM fixes.

Use this skill when baseline NeMo Relay instrumentation exists and the user wants to configure or evaluate adaptive plugin behavior, including telemetry, state, adaptive_hints, tool_parallelism, ACG, hint consumption, or measured rollout.

Top-level workflow skill for USD performance diagnosis and optimization. Use for slow loading, high memory, low FPS, or 'optimize my scene' requests; delegates auth/runtime setup to Phase 0 owners.

Techniques for reducing peak GPU memory in Megatron Bridge — expandable segments, PEFT + SP input re-gather, parallelism resizing, activation recompute, CPU offloading constraints, and common OOM fixes.

Use this skill when baseline NeMo Relay instrumentation exists and the user wants to configure or evaluate adaptive plugin behavior, including telemetry, state, adaptive_hints, tool_parallelism, ACG, hint consumption, or measured rollout.

Top-level workflow skill for USD performance diagnosis and optimization. Use for slow loading, high memory, low FPS, or 'optimize my scene' requests; delegates auth/runtime setup to Phase 0 owners.

Indexed by skills.sh from google/skills

Indexed by skills.sh from google/skills

Indexed by skills.sh from parcadei/continuous-claude-v3

Indexed by skills.sh from aj-geddes/useful-ai-prompts

Indexed by skills.sh from nvidia/skills

Indexed by skills.sh from neo4j-contrib/neo4j-skills

Indexed by skills.sh from nvidia/skills

Indexed by skills.sh from gamedev-skills/awesome-gamedev-agent-skills

Indexed by skills.sh from wshobson/agents

Generate structured answer templates for CIE A-Level Physics (9702) exam questions. Fine-tuned Qwen3-4B LoRA model: question type, given/required, formulae,...

Fine-tune large LLMs with LoRA on limited GPU memory.

Build transformer fine-tuning run plans with task settings, hyperparameters, and model-card outputs. Use for repeatable Hugging Face or PyTorch finetuning wo...

LoRA fine-tuning pipeline for Stable Diffusion on Apple Silicon — dataset prep, training, evaluation with LLM-as-judge scoring. Use when fine-tuning image ge...

Cosmos3-Nano video QA supervised fine-tuning with FSDP parallelism. Use when training or evaluating video question-answering models, fine-tuning Cosmos3-Nano or compatible Cosmos Reason models with SFT/LoRA, or working with Cosmos-RL. Trigger phrases include "fine-tune Cosmos"…

Orchestration skill for NVIDIA Nemotron Speech (Riva) / NeMo ASR domain and language adaptation. Given a goal like "improve/fine-tune ASR for my domain or language", it scopes the task, picks the cheapest sufficient path (word boosting → n-gram LM → fine-tuning), delegates eac…

CLIP vision-language model for image-text retrieval, zero-shot classification, embedding extraction, ONNX export, and TensorRT deployment. Use when fine-tuning or training CLIP, running zero-shot classification, computing image embeddings, or deploying CLIP to ONNX/TensorRT.

Cosmos3-Nano video QA supervised fine-tuning with FSDP parallelism. Use when training or evaluating video question-answering models, fine-tuning Cosmos3-Nano or compatible Cosmos Reason models with SFT/LoRA, or working with Cosmos-RL. Trigger phrases include "fine-tune Cosmos"…

Use when adding, modifying, optimizing, or debugging CuTile autotuning code. Trigger signals: `exhaustive_search` / `replace_hints` / `hints_fn` / `cuda.tile.tune` in code, `autotune` in filenames, or correctness/performance issues in autotuned CuTile kernels. Covers: tune-onc…

Use when adding, modifying, optimizing, or debugging CuTile autotuning code. Trigger signals: `exhaustive_search` / `replace_hints` / `hints_fn` / `cuda.tile.tune` in code, `autotune` in filenames, or correctness/performance issues in autotuned CuTile kernels. Covers: tune-onc…

Vision-language chat: VQA, captioning, image dialogue.

W&B: log ML experiments, sweeps, model registry, dashboards.

Indexed by skills.sh from nvidia/skills

Microsoft Research's agent training framework. Optimizes AI agents with Reinforcement Learning, Automatic Prompt Optimization, and Supervised Fine-tuning. Ze...

Generate GPT API request payloads. Use when building chat completions, embeddings, fine-tuning data, or estimating API costs.

Configure, estimate, and generate LoRA fine-tuning scripts for LLMs. Input: base model name, dataset size, GPU spec. Output: training config, PEFT script, co...

Train high-performance medical LLMs on consumer GPUs using parameter-efficient fine-tuning

MLX Swift LM - Run LLMs and VLMs on Apple Silicon using MLX. Covers local inference, streaming, tool calling, LoRA fine-tuning, and embeddings.

Manage OpenAI files, assistants, vector stores, batches, fine-tuning jobs, and model resources via the OpenAI API. Use this skill when users want to create o...

Reviews AI/ML model supply chains for security risks including model provenance verification, training data lineage, fine-tuning pipeline integrity, inferenc...

Captures learnings about GenAI/LLM configuration, model selection, inference optimization, fine-tuning, RAG pipelines, prompt engineering, multimodal process...

Interactive step-by-step tutor for Solo Hub — guides a human through account setup, model browsing, team management, credits, and fine-tuning (LLM and VLA) u...

Expert LLM architect specializing in large language model architecture, deployment, and optimization. Masters LLM system design, fine-tuning strategies, and...

Use this skill whenever the user mentions Tinker, tinker CLI, training runs, checkpoints, model fine-tuning with Tinker, tinker-cookbook, tinker recipes, or...

Masked Auto-Encoder (MAE) for self-supervised pretraining and fine-tuning. Masks random patches and reconstructs them to learn visual representations; supports pretrain and finetune stages. Use when training, evaluating, exporting, or running inference for a TAO MAE backbone. …

Appeal a parking ticket, penalty charge, or administrative fine with the grounds that actually get appeals granted — not indignation. Use when someone got a...

Provides tools for implementing feedback loops to fine-tune LLM agents using user feedback for continuous personalization and improvement, including training...

Indexed by skills.sh from nomadamas/k-skill

Help the user define a concrete, measurable goal before starting work, especially when they ask to use the goal tool, create a goal, set an objective, clarify success criteria, or turn a fuzzy intention into a quantitative outcome. Use this skill for goal creation and goal ref…

Extended ethos and mental models for Clanky. Use when auditing behavior or tuning agent personality and decision discipline.

Benchmark and compare agent prompts and evaluation results. Use when tuning strategies, evaluating outputs, or comparing configurations.

Alert design: SLOs, noise reduction, routing, severity. Use when tuning pages or defining on-call policy.

Deploy, secure, and troubleshoot Beszel monitoring with Docker agents, alert tuning, and upgrade-safe operations for self-hosted servers

Query, optimize, and administer ClickHouse OLAP databases with schema design, performance tuning, and data ingestion patterns.

Ultimate cost optimization toolkit for OpenClaw/Claude Code. Smart model routing, context compression, heartbeat tuning, usage reports, config generation — s...

Use when designing or tuning exit logic for a Freqtrade strategy on Superior Trade — anything described as ratcheting trailing stop, two-phase exit, ALO-awar...

Generate optimal ESLint configurations based on project type, framework, and team preferences — flat config format, plugin selection, and rule tuning.

Review past Hong Kong IPO decisions, update actual outcomes, export review datasets, and accept or reject tuning suggestions. Use when the user wants to lear...

Manage Hadoop clusters with HDFS operations, YARN job tuning, and distributed processing diagnostics.

Work with HashiCorp Vault using the `vault` CLI for authentication checks, KV secret reads and writes, listing paths, enabling and tuning secrets engines, po...

Evaluate hi-fi and audio gear options, build system recommendations, guide installation and tuning, and analyze used-market pricing/resale value. Use when us...

Set up, troubleshoot, and optimize HomePod and HomeKit audio workflows with reliable Siri control and room-aware playback tuning.

Applies kernel tuning, web server optimization, security hardening, and resource limits to boost Hostinger VPS performance and reduce costs.

A comprehensive set of skills for KES V8/V9, covering the full lifecycle from installation and deployment to performance tuning, data migration, troubleshooting, and application development. Enables AI assistants to provide accurate answers to KES-related questions.

Megasquirt ECU tuning and calibration via TunerStudio, covering VE table, ignition timing, AFR targets, sensor calibration, accel enrichment, boost, launch c...

Use when work targets NVIDIA GPUs for deep learning training, inference, distributed execution, CUDA/Triton kernels, or AI infra tuning. Enforces GPU-aware c...

Manage, optimize, and troubleshoot the OpenClaw memory system — MEMORY.md curation, daily logs (memory/YYYY-MM-DD.md), memory_search tuning, compaction survi...

Manages skill lifecycle and routing using a balanced governance model with status transitions, core pool tuning, and failure-based quarantine.

Audit and improve OpenClaw workspaces across SOUL, AGENTS, TOOLS, USER, MEMORY, and skills for persona tuning, proactive behavior, recall, and workflow fit....

Use when installing, updating, diagnosing, configuring, fixing, tuning, or setting up anything in OpenClaw — gateway not responding, channel silent, model fa...

OpenClaw token cost reduction, performance tuning, and security hardening guide. Use when asked to optimize OpenClaw, reduce token costs, fix missing capabil...

Amazon OpenSearch vector search expert knowledge base. Comprehensive guidance on vector search configuration, cluster tuning, quantization, cost optimization...

Analyze OpenClaw configs and session data to recommend specific, cost-effective token usage and model tuning changes with clear trade-offs and explanations.

PID controller tuning and simulation tool. Use when json pid tasks, csv pid tasks, checking pid status.

Set up and optimize RAG pipelines for large datasets (50K-500K rows) with document chunking, embedding benchmarking, vector indexing, and retrieval tuning.

Use this skill when you need to write, review, optimize, or debug SQL queries. Covers query construction, performance tuning, index strategy, window function...

Deploy MISP threat intelligence platform on any Docker-ready Linux host. Official misp-docker project with automatic MariaDB memory tuning (prevents OOM on s...

Deep vector database workflow—embedding choice, index algorithms, recall/latency trade-offs, hybrid search, filtering, operational tuning, and cost. Use when...

Guide staged text-to-video generation from a rough user idea to ranked video type options, parameter tuning, prompt preview, and final Volcengine Ark video g...

Retrieval and RAG workflow on Volcengine AI stack. Use when users need embedding search, document indexing, top-k retrieval, grounding prompts, or search relevance tuning.

AI video generation workflow on Volcengine. Use when users need text-to-video, image-to-video, generation parameter tuning, or async task troubleshooting for video jobs.

WordPress site health audit, performance optimization, database cleanup, autoload tuning, slow query detection, wp-config management, image size control, fro...