Results for “llm-friendly”

Deploy Meta's LlamaGuard moderation model to filter LLM inputs and outputs across 6 safety categories using HuggingFace, vLLM, or FastAPI.

Train, fine-tune, and deploy LLMs using LitGPT's clean implementations of 20+ architectures like Llama, Gemma, and Phi.

Optimizes LLM inference with NVIDIA TensorRT for maximum throughput and lowest latency on NVIDIA GPUs (A100/H100).

Trains LLMs with SimPO, a reference-free preference optimization method that outperforms DPO, using configurable hyperparameters and workflows for various models and tasks.

Deploy and serve LLMs with high throughput using vLLM's PagedAttention and continuous batching. Supports OpenAI-compatible endpoints, quantization (GPTQ/AWQ/FP8), and tensor parallelism for production inference.

Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag) using standardized prompts and metrics. Supports HuggingFace, vLLM, and API backends.