Skill UI

This guide provides a comprehensive tutorial on running various AI inference tasks using the Together AI API, which is compatible with the OpenAI standard. It demonstrates key functionalities including chat completions, streaming responses, image generation, and creating embeddings. This is ideal for developers who need to test open-source models, compare performance across different LLMs (like Llama or Mixtral), or integrate powerful generative AI features into their applications using Python or Node.js.

This guide provides comprehensive instructions for setting up and authenticating the Together AI SDK. It covers installation using both Python and Node.js, configuring the necessary API key, and verifying connectivity to utilize the OpenAI-compatible API for powerful open-source model inference and fine-tuning tasks. Ideal for developers starting with Together AI.

Provides comprehensive guidance for performance tuning, inference, and fine-tuning using the Together AI platform. This skill supports working with Together AI's OpenAI-compatible API, covering model deployment, utilizing various open-source models (like Llama, Mixtral), and implementing best practices for efficient and cost-effective batch inference.

A comprehensive guide to implementing production-ready patterns for interacting with the Together AI API. This includes a singleton client for seamless OpenAI-compatible integration, robust error wrappers with retry logic, and a fluent request builder. It covers best practices for LLM inference, model setup, and handling diverse response types (chat, embedding, fine-tuning).

This guide provides comprehensive strategies for migrating applications using the Together AI platform. It addresses common issues like model ID deprecations, API schema changes (for fine-tuning or chat completions), and best practices for robust inference and deployment using their OpenAI-compatible API.

This skill provides real-time object detection capabilities by leveraging the Google Coral Edge TPU accelerator. It operates natively within Windows Subsystem for Linux (WSL), allowing high-speed, low-latency inference on live camera frames. By utilizing hardware acceleration, it achieves impressive performance (e.g., ~4ms inference at 320x320), making it ideal for edge computing projects requiring rapid visual analysis and resource efficiency.

This skill provides comprehensive methods and frameworks to detect sophisticated AI attacks, including model extraction, membership inference, and model inversion, which abuse inference APIs. It focuses on monitoring per-principal query volume, input distribution, and confidence exposure. Use it for securing public or partner inference APIs, or for conducting pre-deployment red-teaming to assess the model's extractability and overall security posture against MITRE ATLAS techniques.

A comprehensive guide and skill for deploying large language models (LLMs) locally using the llama.cpp framework and GGUF quantization format. It covers searching the Hugging Face Hub, selecting optimal quantizations (e.g., Q4_K_M), running local servers, and achieving OpenAI-compatible API endpoints. Supports various hardware backends including CPU, Mac Metal, CUDA, and ROCm. Ideal for developers needing offline, private LLM inference.

Utilize zero-shot NER to extract custom entity types (e.g., Drug, Symptom, Device) from complex clinical or biomedical text without requiring any labeled training data or fine-tuning. OpenMed wraps GLiNER/GLiNER2, allowing users to define labels on the fly. The workflow involves building a model index and then running inference via CLI or Python API, making it ideal for rapid prototyping or unstable schemas.

Provides a hardened FastAPI endpoint for OpenMed's core functionalities: clinical Named Entity Recognition (NER), Personally Identifiable Information (PII) extraction, and de-identification. Designed for self-hosted microservices, it handles inference requests, model lifecycle management, and supports dynamic batching, ensuring secure, on-device processing of sensitive clinical data.