223 个（NousResearch 的仓库） · 按人气排序
Explain concepts through detailed, interactive machines drawn from an angled view. The skill produces standalone HTML, SVG, or React components, with moving parts and numerical readouts driven by the same model so readers can explore how the idea works.
中文简介正在准备中。当前显示英语简介。
Access paid APIs using Machine Payments Protocol (MPP), a protocol for request payments. The skill checks payment requirements, selects a compatible wallet or client, submits payment, and helps inspect the API response and receipt.
中文简介正在准备中。当前显示英语简介。
Turn audio and video into text across 99 languages, translate speech into English, and create timestamped subtitles. The skill covers model selection, transcription settings, and batch processing for recordings such as meetings and podcasts.
中文简介正在准备中。当前显示英语简介。
Get guidance for fine-tuning large language models with Axolotl, adapting them through additional training. Use YAML configuration examples, dataset references, and training methods such as LoRA to set up training and investigate problems.
中文简介正在准备中。当前显示英语简介。
Guide language model training with TRL using example answers, preferred response pairs, or reward scores. The skill covers data preparation, method selection, reward model training, evaluation, and adjustments for memory limits or unstable training.
中文简介正在准备中。当前显示英语简介。
Use the hf command-line tool to find, download, and upload models and datasets on Hugging Face Hub. The guide also covers repository management, hosted computing resources, local cache maintenance, and automation options.
中文简介正在准备中。当前显示英语简介。
Use SAM to separate objects from images by marking points or drawing boxes, without training a model for each task. Covers automatic region generation, refining selections, preparing annotation data, and exporting models for deployment.
中文简介正在准备中。当前显示英语简介。
Helps you use Unsloth for language model fine-tuning with LoRA and QLoRA, methods that adapt models with fewer trainable parameters. Supports API questions, implementation, and debugging using reference material drawn from official documentation.
中文简介正在准备中。当前显示英语简介。
Set up vLLM to serve large language models through an OpenAI-compatible API or process prompts in batches. The skill covers deployment, performance monitoring, and quantization, which reduces model memory requirements.
中文简介正在准备中。当前显示英语简介。
Find language models on Hugging Face Hub and run them locally with llama.cpp. The skill helps compare GGUF model files, choose compressed versions that fit available memory, and prepare commands for a local chat server or command-line session.
中文简介正在准备中。当前显示英语简介。
Build question answering and document retrieval workflows with DSPy by defining inputs, outputs, and reusable components. Use training examples and evaluation metrics to automatically optimize prompts and compare results before and after optimization.
中文简介正在准备中。当前显示英语简介。
Generate model responses in defined formats for data extraction and classification. Use JSON schemas, Pydantic data models, regular expressions, or fixed choices to produce structured results that other programs can parse and validate.
中文简介正在准备中。当前显示英语简介。
Find relevant guides and runnable workflows for genomics and computational biology through an index of over 400 skills. Retrieve references or scripts for tasks such as sequence quality checks, gene expression analysis, and reporting.
中文简介正在准备中。当前显示英语简介。
Test a plan before implementation through structured, challenging questions. Work through decisions in dependency order, receive recommendations with each question, and finish with a clear account of agreed choices, open issues, and scope.
中文简介正在准备中。当前显示英语简介。
Support drug discovery research by searching public databases for compound activity, molecular properties, drug interactions, and gene–disease associations. Assess drug-likeness, meaning how molecular properties fit established rules for potential oral medicines.
中文简介正在准备中。当前显示英语简介。
Set up large language model pretraining with TorchTitan and PyTorch. Configure training across multiple GPUs, combine four ways of splitting the workload, enable Float8 calculations on H100 hardware, and manage checkpoints for saving and resuming training.
中文简介正在准备中。当前显示英语简介。
Build landing pages where scrolling advances video and visual effects. The skill guides interviews, visitor journeys, asset preparation, and HTML implementation, then checks screenshots across scroll positions, mobile layouts, and reduced-motion settings.
中文简介正在准备中。当前显示英语简介。
Create a Markdown questionnaire for decisions that depend on someone else's knowledge. Clarify the recipient and needed answers, then organize focused questions so an expert or stakeholder can respond before or during a meeting.
中文简介正在准备中。当前显示英语简介。
Cross-reference public records to investigate companies, contracts, sanctions, property, and litigation. Match names across sources, label confidence levels, and build structured findings linked to original records, with optional statistical analysis of event timing.
中文简介正在准备中。当前显示英语简介。
Turn researched property and rental details into browsable cards in Hermes desktop conversations. Each card can show photos, prices, features, source links, and questions to check, keeping shortlisted properties easy to compare during follow-up discussions.
中文简介正在准备中。当前显示英语简介。
Set up and optimize large language model inference, or response generation, on NVIDIA GPUs. Use quantization, batch processing, and multiple GPUs to improve serving performance, and expose models through a chat completion API.
中文简介正在准备中。当前显示英语简介。
Publish a Cloudflare Worker to a temporary live URL without creating an account first. The skill covers deployment, checking the response, updating the code, and handing over an ownership link before the 60-minute expiry.
中文简介正在准备中。当前显示英语简介。
Search the web, news, images, and videos through DuckDuckGo without an API key. Filter results by region or time and retrieve titles, links, and snippets, including when the usual search tool is unavailable.
中文简介正在准备中。当前显示英语简介。
Retrieve courses and assignments from Canvas LMS, a learning management system. Check due dates, points, submission types, and assignment links, with options to filter active courses and sort assignments by deadline.
中文简介正在准备中。当前显示英语简介。