Search Results for "guard"

LLM Guard is an open-source security toolkit designed to protect large language model applications from various security risks and adversarial attacks. The library acts as a protective layer between users and language models by analyzing inputs and outputs before they reach or leave the model. It includes scanning mechanisms that detect malicious prompts, prompt injection attempts, toxic content, and other harmful inputs that could compromise AI systems.

...Its scope spans input and output safeguards, cybersecurity-focused evaluations, and reference shields that can be inserted at inference time. The project evolves as a hub for safety research artifacts like Llama Guard and Code Shield, along with dataset specs and how-to guides for integrating checks into applications. CyberSecEval, one of its flagship components, provides repeatable evaluations for security risk, including agent-oriented tasks such as automated patching benchmarks. The aim is to make safety practical: ship testable baselines, publish metrics, and provide drop-in implementations that reduce friction for teams adopting Llama. ...

starter-applets is a collection of minimal, sandboxed example “applets” that demonstrate how to compose Gemini-powered microapps (chat widgets, image generation, workflows) that can be embedded in other applications or used standalone. The applets are structured with a focus on simplicity: each presents a prompt input, minimal UI logic, and inline display of the resulting output or widget (e.g. generated text, images). They are built to illustrate best practices (e.g. safety guards, prompt...

NVIDIA-Nemotron-3-Nano-30B-A3B-FP8 is a state-of-the-art large language model developed and released by NVIDIA as part of its Nemotron 3 family, optimized for high-efficiency inference and strong reasoning performance in open AI workloads. It is the post-trained and FP8-quantized variant of the Nemotron 3 Nano model, meaning its weights and activations are represented in 8-bit floating point (FP8) to dramatically reduce memory usage and computational cost while retaining high accuracy. The...

meta-llama/Llama-3.2-1B is a lightweight, instruction-tuned generative language model developed by Meta, optimized for multilingual dialogue, summarization, and retrieval tasks. With 1.23 billion parameters, it offers strong performance in constrained environments like mobile devices, without sacrificing versatility or multilingual support. It is part of the Llama 3.2 family, trained on up to 9 trillion tokens and aligned using supervised fine-tuning, preference optimization, and safety...