Trending

Show HN: GPT-2 implemented using graphics shaders

Suno v4.5

LLMs for Engineering: Teaching Models to Design High Powered Rockets

Llasa: Llama-Based Speech Synthesis

Xiaomi unveils open-source AI reasoning model MiMo

DeepSeek-Prover-V2

I made my AI think harder by making it argue with itself. It works stupidly well

An illustrated guide to automatic sparse differentiation

Finding paths of least action with gradient descent (2023)

LLM GPU calculator for inference and fine-tuning requirements

Granite-4-Tiny-Preview is a 7B A1 MoE

A company gave 1,000 AI agents access to Minecraft — and they built a society

ubergarm/Qwen3-30B-A3B-GGUF 1600 tok/sec PP, 105 tok/sec TG on 3090TI FE 24GB VRAM

[D] Are weight offloading / weight streaming approaches like in Deepseek Zero used frequently in practice? (For enabling inference on disproportionately undersized GPUs)

Build a Text-to-SQL AI Assistant with DeepSeek, LangChain and Streamlit

New TTS/ASR Model that is better that Whisper3-large with fewer paramters

Impressive Qwen 3 30 MoE

[D] DE vs Gen AI/ML

DeepSeek

Save item

DeepSeek-Prover-V2-671B

StepFun

Save item

Step1X-Edit

Task: image-to-image

HiDream.ai

Save item

HiDream-E1-Full

Task: any-to-any

freepik

Save item

F-Lite

Task: text-to-image

NVIDIA

Save item

parakeet-tdt-0.6b-v2

Task: automatic-speech-recognition

DeepSeek

Save item

DeepSeek-V3-0324

NousResearch

Save item

Minos-v1

Task: text-classification

HiDream.ai

Save item

HiDream-I1-Full

Task: text-to-image

NVIDIA

Save item

DAM-3B

Task: image-text-to-text

mikel-brostrom

Save item

boxmot

BoxMOT: pluggable SOTA tracking modules for segmentation, object detection and pose estimation models

NVIDIA

Save item

warp

A Python framework for high performance GPU simulation and graphics

neuraloperator

Save item

neuraloperator

Learning in infinite dimension with neural operators.

steven2358

Save item

awesome-generative-ai

A curated list of modern Generative Artificial Intelligence projects and services

deepspeedai

Save item

DeepSpeed

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

roboflow

Save item

notebooks

This repository offers a comprehensive collection of tutorials on state-of-the-art computer vision models and techniques. Explore everything from foundational architectures like ResNet to cutting-edge models like YOLO11, RT-DETR, SAM 2, Florence-2, PaliGemma 2, and Qwen2.5VL.

BlinkDL

Save item

RWKV-LM

RWKV (pronounced RwaKuv) is an RNN with great LLM performance, which can also be directly trained like a GPT transformer (parallelizable). We are at RWKV-7 "Goose". So it's combining the best of RNN and transformer - great performance, linear time, constant space (no kv-cache), fast training, infinite ctx_len, and free sentence embedding.