社区博客与文章

You can also read this blog in English

Community Articles

Tensor Parallelism (TP) in Transformers: 5 Minutes to Understand

TurkColBERT: A Benchmark of Dense and Late-Interaction Models for Turkish Information Retrieval

Norm-Preserving Biprojected Abliteration

Ellora: Enhancing LLMs with LoRA - Standardized Recipes for Capability Enhancement

An Edge-First Generalized LLM LoRA Fine-Tuning Framework for Heterogeneous GPUs

Uncensor any LLM with abliteration

Code a simple RAG from scratch

AI Energy Score v2: Refreshed Leaderboard, now with Reasoning 🧠

Building Jobly: Semantic Job Matching with RAG and Vector Embeddings

Making Model Tuning Accessible: This is what we built observing 100s of users tune models!

DeepFabric: Generate, Train and Evaluate with Datasets curated for Model Behavior Training.

Gemini-3 Benchmarkathon

Engineering Notes: Training a LoRA for Z-Image Turbo with the Ostris AI Toolkit

Mastering Tensor Dimensions in Transformers

Curating datasets directly on the Hub

Building and evaluating Multimodal Rerankers

KV Caching Explained: Optimizing Transformer Inference Efficiency

A Review on the Evolvement of Load Balancing Strategy in MoE LLMs: Pitfalls and Lessons

From GRPO to DAPO and GSPO: What, Why, and How

GSMA Open-Telco LLM Benchmarks 2.0: The first dedicated LLM Evaluation for Telecoms

researchcommunity

中国 AI 出海现状概述

2024年10月3日

顶点着色网格转换为 UV 映射的纹理化网格

2024年9月30日

multimodalon-devicellm

现在 Llama 具备视觉能力并可以在你的设备上运行 - 欢迎使用 Llama 3.2

+3

2024年9月25日

videodatasetsmultimodal

揭秘 FineVideo 数据集构建的背后的秘密

+2

2024年9月23日

researchcommunity

Hugging Face 论文平台 Daily Papers 功能全解析

2024年9月23日

inteloptimumquantization

使用 Optimum-Intel 和 OpenVINO GenAI 优化和部署模型

+3

2024年9月20日

nlpresearchcommunity

Fine-tuning LLMs to 1.58bit: extreme quantization made easy

+2

2024年9月18日

datasetssqlduckdb

为数据集而生的 SQL 控制台

2024年9月17日

Accelerate 1.0.0

2024年9月13日

paddingpackingFlash Attention 2

通过打包 Flash Attention 来提升 Hugging Face 训练效率

+2

2024年8月21日

long-contextinfini-attentionmemory-compression

一次失败的实验——无限注意力，我们为什么坚持实验

2024年8月14日

guidecommunityggml

ggml 简介

2024年8月13日

nlpcommunityresearch

Falcon Mamba: 首个高效的无注意力机制 7B 模型

+2

2024年8月12日

LLMnlpcommunity

对 LLM 工具使用进行统一

2024年8月12日

Community Articles

NEW Articles from Team or Enterprise organizations will get promoted to the main section.

Tensor Parallelism (TP) in Transformers: 5 Minutes to Understand

TurkColBERT: A Benchmark of Dense and Late-Interaction Models for Turkish Information Retrieval

Norm-Preserving Biprojected Abliteration

Ellora: Enhancing LLMs with LoRA - Standardized Recipes for Capability Enhancement

An Edge-First Generalized LLM LoRA Fine-Tuning Framework for Heterogeneous GPUs

Uncensor any LLM with abliteration

Code a simple RAG from scratch

AI Energy Score v2: Refreshed Leaderboard, now with Reasoning 🧠

Building Jobly: Semantic Job Matching with RAG and Vector Embeddings

Making Model Tuning Accessible: This is what we built observing 100s of users tune models!

DeepFabric: Generate, Train and Evaluate with Datasets curated for Model Behavior Training.

Gemini-3 Benchmarkathon

Engineering Notes: Training a LoRA for Z-Image Turbo with the Ostris AI Toolkit

Mastering Tensor Dimensions in Transformers

Curating datasets directly on the Hub

Building and evaluating Multimodal Rerankers

KV Caching Explained: Optimizing Transformer Inference Efficiency

A Review on the Evolvement of Load Balancing Strategy in MoE LLMs: Pitfalls and Lessons

From GRPO to DAPO and GSPO: What, Why, and How

GSMA Open-Telco LLM Benchmarks 2.0: The first dedicated LLM Evaluation for Telecoms

View all articles