jon123boss/fast-attnres: Fast BF16 PyTorch training operators for standard and low-rank Attention Residuals
Fast BF16 PyTorch training operators for standard and low-rank Attention Residuals
Fast BF16 PyTorch training operators for standard and low-rank Attention Residuals
Physics-Informed Neural Network for a damped mass-spring oscillator. Includes ablation experiments on IC enforcement, loss weighting, and sampling strategy, with a hard constraint ...
Turn any AI agent into an AI Scientist. The #1 Agent Skills library for science, used by 190,000+ scientists worldwide. 165 ready-to-use validated skills plus 100+ scientific datab...
易创:纯本地、开源的 AI 网文写作桌面软件,支持小说创作、AI 辅助写作、BYOK 与自定义提示词。
想做真人口播,又不必自己对着镜头念。上传一段真人出镜视频,写好字幕,本机配音并对口型,合成竖屏 MP4。密钥只留在浏览器里。
SecretShield is a cybersecurity tool that uses machine learning and pattern-based analysis to detect potential secrets and sensitive credentials in source code, helping developers ...
Football transfer intelligence platform that combines player performance, historical transfer data, and machine learning to estimate transfer values.
Benchmarking models as runtime Controllers for Loop Engineering
Continual learning infra for self-improving agents
Verified reverse engineering: AI RE grounded on deterministic tools - results checked against the binary, not hallucinated.
Official implementation of “Enabling KV Caching of Shared Prefix for Diffusion Language Models” (EMNLP 2026 Main Conference)
LLM aberto e reprodutível para o Português do Brasil An Open, Reproducible Brazilian-Portuguese Language Model
Complete AI Engineering course for beginners covering LLMs, Generative AI, RAG, LangChain, AI Agents, embeddings, vector databases, and hands-on projects. Companion code and Jupyte...
Rectified flow diffusion transformer sampler & linear vector field (Flux.1 DiT style)
Python library for memory-constrained activation checkpoint optimization, recomputation scheduling, and GPU training performance analysis.
Blueprint-style visual editor for PyTorch — draw neural architectures, compile to torch.nn.Module, train, trace, and infer locally.
A ComfyUI custom node that routes MiniMax H3 diffusion transformer attention through FastVideo's Video Sparse Attention (VSA). Combined with FastVideo's 4-step distillation LoRA, i...
AI 短剧生产总控 / 短剧skill:专业导演与分镜、MiniMax H3、CompShare、声音资产、粗剪与 QC,默认 16:9。
[ECCV 2026] DualDiff3D: Dual Structure-Appearance Diffusion Priors for Reliability-Enhanced 3D Gaussian Splatting
A web-based Generative AI Image Generation Studio built with Flask, Hugging Face Inference Providers, and FLUX.1-dev.
Generate a YC startup name before someone else does.
Privacy-first AI coding activity for your GitHub profile.