Inferflow is an efficient and highly configurable inference engine for large language models (LLMs).

MiniMax-H3 multi-GPU parallel inference for ComfyUI | 多卡并行加速节点:2-8 GPU Ulysses sequence parallel, bit-identical video+audio generation

A script for PyTorch multi-GPU multi-process testing

Distributed Reinforcement Learning for LLM Fine-Tuning with multi-GPU utilization

ComfyUI Qwen-Image 2.1 multi-GPU inference: Ulysses sequence parallelism for 1/2/4/8 GPUs, T2I, image editing, INT8/FP8, and distributed prefix cache.

A reproducible Docker-based pipeline for running machine learning experiments with GPU support. This repository provides pre-configured Docker images, environment files, and scripts for: - Setting up GPU-enabled containers - Running training and inference - Managing environments reproducibly

Nemotron-3-Ultra-550B, Gemma4-31B on 8xA100 40GB GPUs as production AWS SageMaker endpoint with tensor-parallel inference, Gemma4-12B-it on Lambda GPU, Megatron-Bridge w vLLM or SGLang,

Kaggle T4×2 demo with two independent INT8 Qwen3-Embedding-8B replicas, FastAPI, FAISS, EN/VI Wikidata search, and evidence-backed qualification.

Lightweight terminal launcher and auto-optimizer for llm models using llama.cpp with hardware detection, tensor sharding, benchmarks, presets, context tuning, and OpenAI-compatible serving.