# transformers Published articles for transformers. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## HeyGen x Google Cloud: Bringing Avatar IV to TPUs DevFeed: [HeyGen x Google Cloud: Bringing Avatar IV to TPUs]() Original publisher: [Read original article]() Published: 2026-09-12T11:04:33.891311Z Content type: article Language: en Sources: [Google Developers Blog]() Topics: [Google](), [Cloud](), [Artificial Intelligence](), [Optimization](), [AI Infrastructure](), [Transformer](), [PyTorch](), [Compiler](), [Streaming](), [real-time](), [Hardware](), [API]() Tags: [ai](), [ai-infrastructure](), [api](), [cloud](), [code](), [compiler](), [engineering](), [generation](), [google](), [hardware](), [model](), [optimization](), [performance](), [pytorch](), [real-time](), [streaming](), [time](), [tpu](), [transformers](), [video]() ### AI overview HeyGen and Google Cloud describe porting the 18B+ parameter Avatar IV talking-head video generation pipeline to an eight-chip Trillium TPU host. Using torchax, JAX, XLA, FSDP sharding, Ulysses sequence parallelism, and custom Pallas kernels, the team improved performance by 1.86x for real-time chunked streaming while preserving output quality through strict quality gates. ### Source excerpt HeyGen ported their 18B+ parameter Avatar IV video generation model to Google Cloud's Trillium (v6e) TPUs via torchax and XLA, utilizing FSDP and Ulysses sequence parallelism across an eight-chip mesh. To achieve a 1.86x speedup for real-time streaming, the engineering team pipelined exposed all-to-all collectives, aligned sparse attention block sizes to eliminate mask padding, and bypassed softmax serial dependencies using a precomputed Cauchy-Schwarz upper bound. These custom Pallas kernel and compiler optimizations were deployed only after passing rigorous two-tier quality gates to guarantee byte-identical or mathematically equivalent pixel outputs. ## Открываем претрейн Alice AI Search: как устроена модель быстрых ответов Алисы на Поиске DevFeed: [Открываем претрейн Alice AI Search: как устроена модель быстрых ответов Алисы на Поиске]() Original publisher: [Read original article]() Author: pet67 (Яндекс) Published: 2026-09-11T06:05:13Z Content type: article Language: ru Sources: [Яндекс - Как мы делаем Яндекс / Статьи]() Topics: [Artificial Intelligence](), [AI search](), [Transformers](), [Mixture of Experts (MoE)](), [Large Language Model](), [hugging face]() Tags: [agentic](), [ai](), [ai-search](), [alice-ai](), [hugging-face](), [llm](), [ml](), [moe](), [rl](), [tag-178bc8f01f24](), [tag-4004cf5948d3](), [tag-61cd5a476b1d](), [tag-d89cae10e887](), [tag-e6d9cc1f0757](), [transformers]() ### AI overview This developer article explains the Alice AI Search pipeline for generating fast answers, including its search and context-processing stages, shorter information contexts, a sparse Mixture-of-Experts architecture combined with an Encoder-Decoder, and online reinforcement learning from user behavior signals. It also announces the open release of the Alice AI-T5-35B-A0.6B Base model, with external inference available through Hugging Face Transformers while optimized production inference remains internal to Yandex. ### Source excerpt Быстрый ответ Алисы AI -- это самый массовый генеративный продукт Яндекса и первое соприкосновение с Алисой для пользователей Поиска. Даже в час пиковой нагрузки пользователь должен получить лаконичный ответ за считаные секунды. Для этого мы, команда Alice AI Search, адаптируем весь пайплайн быстрых ответов -- от собственного претрейна с кастомной архитектурой до онлайн-rl-обучения на поведенческие сигналы пользователей. В статье разберём, как устроен генеративный ответ в Поиске, и расскажем про основные улучшения июньского релиза: как мы ускорили ответы за счёт коротких инфоконтекстов, зачем совместили Encoder-Decoder с разреженной MoE-архитектурой и как обучение на реальных пользовательских сигналах повлияло на качество и использование продукта. Кроме того, мы выложили в открытый доступ обученную с нуля модель Alice AI-T5-35B-A0.6B Base с тем ограничением, что внешним пользователям доступен инференс через Hugging Face Transformers, а оптимизированный production-инференс пока доступен только внутри Яндекса. Читать далее ## Run AI in the Browser: A Practical Guide to Transformers.js DevFeed: [Run AI in the Browser: A Practical Guide to Transformers.js]() Original publisher: [Read original article]() Author: Freek Van der Herten (freek@spatie.be) Published: 2026-09-09T14:50:26Z Content type: tutorial Language: en Sources: [freek.dev - all blogposts]() Topics: [transformers.js](), [browser](), [Artificial Intelligence](), [AI Models](), [Back end](), [API keys]() Tags: [ai](), [ai-models](), [api-keys](), [article](), [backend](), [browser](), [guide](), [internet](), [javascript](), [js](), [machine-learning](), [practical](), [run](), [transformers](), [transformers-js]() ### AI overview A practical guide to Transformers.js, covering how it runs AI models directly in the browser, available models, and the trade-offs of client-side AI compared with traditional AI providers. ### Source excerpt Transformers.js lets you run AI models directly in the browser without a backend, API keys, or an internet connection after the model is cached. The article explores how it works, which models are available, and the trade-offs of client-side AI compared to traditional AI providers. Read more ## NeoMME: an efficient Multimodal-native and Multilingual Encoder DevFeed: [NeoMME: an efficient Multimodal-native and Multilingual Encoder]() Original publisher: [Read original article]() Author: Tony Wu; Aurélien Lac Published: 2026-09-03T13:13:48Z Content type: article Language: en Sources: [Hugging Face - Blog]() Topics: [vlm](), [Fine-tuning](), [Large Language Model](), [GPU]() Tags: [apache](), [architecture](), [diffusion](), [embeddings](), [fine-tuning](), [gpu](), [hugging-face](), [multimodal](), [quantization](), [retrieval](), [training](), [transformers](), [vector](), [vision](), [vlm]() ### AI overview NeoMME is a family of multilingual multimodal encoders trained from scratch with a masked discrete-diffusion objective. It uses one bidirectional Transformer for text tokens and image patches, and is fine-tuned for visual document retrieval with dense and late-interaction embeddings. ### Source excerpt We introduce NeoMME, a family of 260M and 800M multilingual multimodal encoders. Unlike many generative visual language models, NeoMME does not use a separate pretrained vision tower or a causal language model. A single bidirectional Transformer processes both text tokens and raw image patches, and we train the entire model from scratch with a masked discrete-diffusion objective. We fine-tuned NeoMME for visual document retrieval using ColPali's page-image approach. ## Multi-Vector (Late Interaction) Embedding Models with Sentence Transformers DevFeed: [Multi-Vector (Late Interaction) Embedding Models with Sentence Transformers]() Original publisher: [Read original article]() Author: Tom Aarsen; Antoine Chaffin; Raphael Sourty Published: 2026-08-18T00:00:00Z Content type: article Language: en Sources: [Hugging Face - Blog]() Topics: [sentence-transformers](), [Embeddings](), [Open Source]() Tags: [community](), [embedding](), [embeddings](), [guide](), [multimodal](), [nlp](), [open-source](), [retrieval](), [search](), [sentence-transformers](), [transformers](), [vector]() ### AI overview This article explains multi-vector, or late-interaction, embedding models with Sentence Transformers. It covers token-level representations, MaxSim scoring, retrieval over text and page images, integration with search systems, and index-size tradeoffs. ### Source excerpt We're on a journey to advance and democratize artificial intelligence through open source and open science. ## This Week In AI Research (🗓 9-16 July 26) DevFeed: [This Week In AI Research (🗓 9-16 July 26)]() Original publisher: [Read original article]() Author: Dr. Ashish Bamania Published: 2026-07-22T09:14:26Z Content type: article Language: en Sources: [Into AI]() Topics: [Artificial Intelligence](), [AI Research](), [releases](), [Large Language Model](), [Mixture of Experts (MoE)](), [long-context](), [Open Source Models & Datasets](), [Transformers]() Tags: [ai](), [architecture](), [benchmarks](), [context-window](), [llms](), [models](), [moe](), [open](), [releases](), [research](), [transformers]() ### AI overview A weekly roundup of AI research papers and releases, covering Kimi K3, Expanded Hyper-Connections, VideoChat3, and other developments. It describes Kimi K3's architecture, context window, benchmark performance, and limitations, and summarizes xHC's reported efficiency improvements. ### Source excerpt The top 10 AI research papers and releases this week (Kimi K3, Inkling, WanSong v1.0, Bonsai 27B, and many more) ## Welcome Inkling by Thinking Machines DevFeed: [Welcome Inkling by Thinking Machines]() Original publisher: [Read original article]() Author: ben burtenshaw; merve; Pedro Cuenca; Aritra Roy Gosthipaty; Andres Marafioti Published: 2026-07-15T00:00:00Z Content type: article Language: en Sources: [Hugging Face - Blog]() Topics: [Large Language Model](), [Mixture of Experts (MoE)](), [multimodal](), [hugging face](), [Inference](), [Deployment](), [Fine-tuning](), [Open Source](), [sglang]() Tags: [announcement](), [artificial-intelligence](), [audio](), [context-window](), [deployment](), [fine-tuning](), [generation](), [inference](), [inference-endpoints](), [llama-cpp](), [llms](), [mixture-of-experts](), [mixture-of-experts-moe](), [moe](), [multimodal](), [open-source](), [sglang](), [transformers](), [vllm]() ### AI overview Thinking Machines Lab's Inkling is presented as a large open multimodal language model that accepts image, text, and audio inputs. The article covers its mixture-of-experts architecture, million-token context window, reasoning across modalities, fine-tuning use cases, model variants, and deployment through Hugging Face Inference Endpoints and inference frameworks. ### Source excerpt We're on a journey to advance and democratize artificial intelligence through open source and open science. ## Native-speed vLLM transformers modeling backend DevFeed: [Native-speed vLLM transformers modeling backend]() Original publisher: [Read original article]() Author: Harry Mellor; Lysandre Published: 2026-07-08T00:00:00Z Content type: article Language: en Sources: [Hugging Face - Blog]() Topics: [vllm](), [Inference Performance](), [Inference](), [LLMs](), [Machine learning](), [Open Source]() Tags: [architecture](), [batching](), [gpu](), [inference](), [inference-performance](), [llama](), [llama-cpp](), [llms](), [mixture-of-experts](), [mlx](), [open-source](), [optimization](), [sglang](), [transformers](), [vllm]() ### AI overview This article describes the transformers modeling backend for vLLM and its role in serving Hugging Face models without requiring model authors to port their implementations. It explains how transformers supplies modeling code while vLLM contributes optimized inference techniques, and outlines comparisons with vLLM's native implementations across several Qwen3 model configurations. It also discusses inference-performance factors including GPU parallelization, compilation, fused kernels, and custom attention kernels. ### Source excerpt We're on a journey to advance and democratize artificial intelligence through open source and open science. ## Token efficiency: getting more signal into the context window DevFeed: [Token efficiency: getting more signal into the context window]() Original publisher: [Read original article]() Author: Jim Allen Wallace Published: 2026-07-01T00:00:00Z Content type: article Language: en Sources: [Redis Blog]() Topics: [long-context](), [Transformer](), [Large Language Model](), [Retrieval Augmented Generation (RAG)]() Tags: [agentic](), [context-window](), [efficiency](), [generation](), [hallucinations](), [llms](), [rag](), [reasoning](), [tech-de](), [tokens](), [transformers]() ### AI overview This guide explains why adding more context can worsen model performance. It examines token efficiency, the "lost in the middle" effect, context rot, attention limits in transformers, and the gap between advertised and effective context windows, with implications for RAG and agentic systems. ### Source excerpt You've probably hit this counterintuitive moment: you give your model more context to work with, expecting better answers, and the answers get worse. More tokens were supposed to mean more information, more grounding, fewer hallucinations. Instead, yo... ## Why intent prediction needs more than an LLM DevFeed: [Why intent prediction needs more than an LLM]() Original publisher: [Read original article]() Author: Phoebe Sajor Published: 2026-06-30T07:40:00Z Content type: article Language: en Sources: [Stack Overflow Blog]() Topics: [foundation-models](), [personalization](), [Graphs](), [Large Language Model](), [data](), [Deep neural networks]() Tags: [agentic-ai](), [ai](), [data](), [foundation-models](), [llm](), [llms](), [math](), [personalization](), [podcast](), [se-stackoverflow](), [se-tech](), [transformers]() ### AI overview Ryan Donovan interviews Frank Portman, CTO at Yobi, about why large language models are not the right inductive bias for predicting human intent and behavior. The discussion covers Yobi's behavioral foundation models, which use transformers and graph neural networks to support large-scale personalization while keeping consumer data private. ### Source excerpt Ryan sits down with Frank Portman, CTO at Yobi, to talk about why next-token prediction, though great for language, isn't the right inductive bias for forecasting human behavior. They discuss how Yobi builds a "foundation model of behavior" using transformers and graph neural networks instead of chat-style LLMs, and what it takes to run millions of personalization decisions per second while keeping consumer data private. ## Using Transformer Architectures for Motion Planning in Autonomous Vehicles DevFeed: [Using Transformer Architectures for Motion Planning in Autonomous Vehicles]() Original publisher: [Read original article]() Author: msporyshev (Яндекс) Published: 2026-06-30T07:04:04Z Content type: article Language: ru Sources: [Яндекс - Как мы делаем Яндекс / Статьи]() Topics: [Machine learning]() Tags: [machine-learning](), [ml](), [reinforcement-learning](), [self-driving](), [tag-4004cf5948d3](), [transformers]() ### AI overview The article explains how Yandex uses transformer-based architectures for motion planning in autonomous vehicles. It covers the difference between open-loop and closed-loop evaluation, the limits of prediction metrics, and the safety-focused validation process from machine-learning experiments through supervised real-vehicle testing. ### Source excerpt Трансформеры уже умеют писать код, генерировать тексты и рисовать картины. Но могут ли они управлять автономным автомобилем в реальных городских условиях, среди людей и других машин? Меня зовут Максим, я руковожу службой поведения и предсказания движения в Автономном транспорте Яндекса. Отвечаю за задачу Motion Planning -- ту часть системы, которая решает, как именно должен двигаться автомобиль в следующие секунды. В этой статье я расскажу, как мы используем архитектуры на основе трансформеров в планировании движения и почему это сильно отличается от привычных задач генеративных моделей. Мы пройдём путь от первых ML-экспериментов до регулярных испытаний на реальных машинах. Разберём, чем Open Loop отличается от Closed Loop, почему качество предсказаний не определяет качество вождения и почему безопасность для нас важнее любой архитектуры. Читать далее ## LLMs: The Essential Guide DevFeed: [LLMs: The Essential Guide]() Original publisher: [Read original article]() Author: Nikki Siapno Published: 2026-06-26T14:44:38Z Content type: tutorial Language: en Sources: [Level Up Coding System Design Newsletter]() Topics: [Large Language Model](), [systems](), [tokenization](), [Transformer](), [distributed-systems](), [Inference](), [Embeddings]() Tags: [embeddings](), [guide](), [inference](), [latency](), [llm](), [llms](), [tokenization](), [transformers]() ### AI overview This guide explains LLMs as distributed systems with layers including models, training pipelines, retrieval, inference infrastructure, safety layers, and product experience. It introduces how tokenization and Transformer architectures work, and connects system layers to debugging, scaling, latency, hallucinations, context loss, and cost. ### Source excerpt Most engineers only see the API layer. Here's what's happening underneath. ## Accelerating Transformers Fine-Tuning with NVIDIA NeMo AutoModel DevFeed: [Accelerating Transformers Fine-Tuning with NVIDIA NeMo AutoModel]() Original publisher: [Read original article]() Author: Adil Asif; Alexandros Koumparoulis; Wenwen Gao; Sylendran Arunagiri; David Messina; Bernard Nguyen Published: 2026-06-24T16:00:13Z Content type: article Language: en Sources: [Hugging Face - Blog]() Topics: [NeMo](), [Fine-tuning](), [AI Models](), [Mixture of Experts (MoE)](), [distributed-training](), [Generative AI](), [Artificial Intelligence](), [GPU](), [PyTorch](), [vllm](), [sglang]() Tags: [ai](), [ai-models](), [building](), [compute](), [distributed-training](), [fine-tuning](), [framework](), [generative](), [generative-ai](), [gpu](), [infrastructure](), [kernels](), [memory](), [model](), [moe](), [nemo](), [nemotron](), [nvidia](), [open](), [transformers]() ### AI overview This article explains how NVIDIA NeMo AutoModel accelerates fine-tuning of mixture-of-experts models by extending Transformers v5 with Expert Parallelism, DeepEP fused all-to-all dispatch, and TransformerEngine kernels. It describes API compatibility, distributed execution, dynamic weight loading, and reported gains of 3.4-3.7x higher training throughput and 29-32% lower GPU memory use. ### Source excerpt NVIDIA NeMo AutoModel is an open library part of the NVIDIA NeMo framework for building custom generative AI models at scale. NeMo AutoModel builds cleanly on top of v5, adding Expert Parallelism, DeepEP fused all-to-all dispatch, and TransformerEngine kernels, and it leans on v5's dynamic weight loading to bring those optimizations to a broad and growing set of model families. ## Experimenting with the proposed Cross-Origin Storage API in Transformers.js DevFeed: [Experimenting with the proposed Cross-Origin Storage API in Transformers.js]() Original publisher: [Read original article]() Author: Thomas Steiner Published: 2026-06-23T00:00:00Z Content type: tutorial Language: en Sources: [Hugging Face - Blog]() Topics: [AI, ML & Data Engineering]() Tags: [ai-models](), [announcement](), [api](), [artificial-intelligence](), [asr](), [browser](), [cache](), [chrome](), [chrome-extension](), [cos](), [cross-origin-storage](), [inference](), [javascript](), [on-device](), [open-source](), [storage](), [transformers](), [transformers-js](), [wasm](), [web](), [web-apps](), [web-developers](), [web-standards](), [webassembly]() ### AI overview A tutorial that uses Transformers.js browser inference examples to examine duplicate model and WebAssembly downloads across origins and the proposed Cross-Origin Storage API. ### Source excerpt We're on a journey to advance and democratize artificial intelligence through open source and open science. ## PP-OCRv6 on Hugging Face: 50-Language OCR from 1.5M to 34.5M Parameters DevFeed: [PP-OCRv6 on Hugging Face: 50-Language OCR from 1.5M to 34.5M Parameters]() Original publisher: [Read original article]() Author: AlexZhang; Cuicheng; Jun Zhang; Manhui Lin; Yue Zhang; Leo-Q8; Yubo; Yi Liu Published: 2026-06-22T13:18:56Z Content type: article Language: en Sources: [Hugging Face - Blog]() Topics: [hugging face](), [Inference](), [Deployment](), [model-deployment]() Tags: [deployment](), [images](), [inference](), [model](), [production](), [transformers]() ### AI overview PP-OCRv6 is PaddleOCR's lightweight OCR model family for multilingual text detection and recognition across documents, screenshots, displays, industrial labels, and scene text. It offers tiny, small, and medium tiers ranging from 1.5M to 34.5M parameters, with deployment options including PaddlePaddle, Transformers, and ONNX Runtime. ### Source excerpt Evaluate PP-OCRv6 online, then integrate lightweight, production-ready OCR with PaddlePaddle, Transformers, or ONNX Runtime backend. PP-OCRv6 is the latest generation of PaddleOCR's universal OCR model family. It is designed for real-world text detection and recognition across documents, screenshots, multilingual images, digital displays, industrial labels, and scene text. The model family scales from 1.5M to 34.5M parameters, with three tiers: tiny, small, and medium. ## Is it agentic enough? Benchmarking open models on your own tooling DevFeed: [Is it agentic enough? Benchmarking open models on your own tooling]() Original publisher: [Read original article]() Author: Lysandre; Nathan Habib; Pedro Cuenca Published: 2026-06-18T00:00:00Z Content type: article Language: en Sources: [Hugging Face - Blog]() Topics: [benchmarking](), [Tooling](), [LLM evaluation / benchmarking](), [AI-assisted coding](), [jobs](), [hugging face](), [Open Source]() Tags: [agents](), [benchmark](), [benchmarking](), [cli](), [code](), [coding](), [developers](), [development](), [evaluation](), [hugging-face](), [jobs](), [open-source](), [tooling](), [transformers]() ### AI overview A benchmark for evaluating how effectively open models use developer tooling, measuring the work required to reach an answer across models, library revisions, and tasks. The article presents an agent-focused harness running on open models with workloads distributed through Hugging Face Jobs. ### Source excerpt We're on a journey to advance and democratize artificial intelligence through open source and open science. ## Beyond LoRA: Can you beat the most popular fine-tuning technique? DevFeed: [Beyond LoRA: Can you beat the most popular fine-tuning technique?]() Original publisher: [Read original article]() Author: Benjamin Bossan; Sayak Paul; Marian Tietz; Kashif Rasul Published: 2026-06-18T00:00:00Z Content type: article Language: en Sources: [Hugging Face - Blog]() Topics: [AI, ML & Data Engineering](), [SDKs]() Tags: [artificial-intelligence](), [community](), [diffusers](), [fine-tuning](), [guide](), [lora](), [nlp](), [peft](), [quantization](), [stable-diffusion](), [training](), [transformers]() ### AI overview The article examines parameter-efficient fine-tuning (PEFT) for open models, focusing on whether LoRA is always the best technique. It explains how PEFT reduces memory requirements, can enable fine-tuning of quantized models, and offers a unified API for multiple techniques. ### Source excerpt We're on a journey to advance and democratize artificial intelligence through open source and open science. ## Making User-Sequence Data More Cost-Efficient, Faster, and Easier to Use DevFeed: [Making User-Sequence Data More Cost-Efficient, Faster, and Easier to Use]() Original publisher: [Read original article]() Author: Pinterest Engineering Published: 2026-05-21T16:01:00Z Content type: article Language: en Sources: [Pinterest Engineering Blog - Medium]() Topics: [data](), [recommendation systems](), [datasets](), [Machine learning](), [Embeddings](), [Feature Engineering](), [Inference](), [Latency](), [systems]() Tags: [data](), [data-infrastructure](), [datasets](), [embeddings](), [engineering](), [feature-engineering](), [inference](), [latency](), [machine-learning](), [pinterest](), [production](), [recommendation-system](), [recommendation-systems](), [retrieval](), [systems](), [train](), [transformers]() ### AI overview Pinterest describes a redesign of its user-sequence platform for ranking, retrieval, and recommendation workloads. The article explains how enriched event sequences support training datasets, offline analysis, online inference, and latency-sensitive production use cases, with goals of reducing cost, improving extensibility, and simplifying debugging. ### Source excerpt Authors (listed alphabetically) Ads Feature Engineering Infra team: Ajay Venkatakrishnan, Le Zhang Core ML Infra team: Eric Shang, Pihui Wei ML Data team: Connor Votroubek, Yi He User Understanding team: Camilo Munoz, Simin Li If you work on ranking, retrieval, or recommendation systems, you've probably asked for some version of the same thing: "Give me the last N meaningful actions this user took, with the right enrichments, in a format that's easy to train and serve ML models." On paper, that sounds simple. In practice, "user sequences" often become one of the most expensive and fragile parts of the ML data stack. They end up powering everything from training datasets to offline analysis and online inference, so they need to be fresh and complete at the same time. They must remain consistent as you add new events and enrichments. And they have to do all of this while serving latency-sensitive production workloads. This article walks through how we redesigned our user-sequence platform to make these sequences cheaper to run, faster to extend, and easier to debug, while still supporting demanding production use cases. What We Mean by "User Sequence" In this context, a user sequence is an ordered list of recent, relevant events for a user, along with the enrichments (signals) attached to each event. Here, enrichments mean all the extra signals we attach to raw events, so they're useful for models: embeddings (for example, Pin or query representations), contextual features (such as surface, device, or country), and derived attributes or counters that describe how the user interacted with a piece of content over time. A concrete example helps. Imagine a sequence made up of the last 500 engagements a user had with Pinterest Pins. Each event in that sequence might carry a timestamp, an action type, the surface where the action occurred, and a handful of embedding features or categorical attributes. As a data primitive, user sequences are powerful. They capture temporal b ## PaddleOCR 3.5: Running OCR and Document Parsing Tasks with a Transformers Backend DevFeed: [PaddleOCR 3.5: Running OCR and Document Parsing Tasks with a Transformers Backend]() Original publisher: [Read original article]() Author: AlexZhang; Cuicheng; Jun Zhang; Manhui Lin; Yue Zhang Published: 2026-05-18T15:12:46Z Content type: release Language: en Sources: [Hugging Face - Blog]() Topics: [document ai](), [Inference](), [hugging face](), [PyTorch](), [Retrieval Augmented Generation (RAG)]() Tags: [document-ai](), [hugging-face](), [inference](), [ocr](), [python](), [pytorch](), [rag](), [rocm](), [transformers]() ### AI overview PaddleOCR 3.5 adds Transformers as a supported inference backend for OCR and document parsing models, while PaddleOCR continues to manage the underlying pipelines. The release simplifies integration with Hugging Face-centered environments and downstream document workflows such as RAG, search, analytics, and automation. ### Source excerpt PaddleOCR continues to provide OCR model series such as PP-OCRv5 and document parsing model series such as PaddleOCR-VL 1.5, while Transformers becomes one of the supported backends for running them. Try the live demo on Hugging Face Spaces: PaddleOCR 3.5 introduces a more flexible inference-engine interface. Developers can select the backend through the parameter and pass backend-specific options through . ## A guide to Transformer architecture in modern language models DevFeed: [A guide to Transformer architecture in modern language models]() Original publisher: [Read original article]() Author: Ashish Pratap Singh Published: 2026-05-14T04:15:11Z Content type: tutorial Language: en Sources: [AlgoMaster Newsletter]() Topics: [Transformer architecture](), [LLMs](), [Transformer]() Tags: [architecture-pattern](), [better](), [deep-dive](), [layer](), [llms](), [model](), [parallel](), [performance](), [semantics](), [sequence](), [syntax](), [tokens](), [transformer-architecture](), [transformers]() ### AI overview This tutorial explains the Transformer architecture, including its original encoder-decoder design for translation and the decoder-only variant used for modern language generation. It describes decoder components such as masked multi-head self-attention, feed-forward networks, layer normalization, and residual connections, and introduces the Pre-LayerNorm pattern. ### Source excerpt A single 2017 research paper changed the future of AI forever and gave rise to multiple unicorn companies. ## Unlocking asynchronicity in continuous batching DevFeed: [Unlocking asynchronicity in continuous batching]() Original publisher: [Read original article]() Author: Rémi Ouazan Reboul; Pedro Cuenca; Aritra Roy Gosthipaty Published: 2026-05-14T00:00:00Z Content type: article Language: en Sources: [Hugging Face - Blog]() Topics: [Inference](), [Large Language Model](), [GPU](), [cpu](), [Cache](), [inference-endpoints]() Tags: [compute](), [cpu](), [cuda](), [gpu](), [inference](), [llm](), [optimization](), [performance](), [pytorch](), [transformers]() ### AI overview This article explains how asynchronous batching improves LLM inference by allowing CPU batch preparation and GPU computation to run in parallel, reducing idle time and improving GPU utilization. ### Source excerpt We're on a journey to advance and democratize artificial intelligence through open source and open science. ## How to Use Transformers.js in a Chrome Extension DevFeed: [How to Use Transformers.js in a Chrome Extension]() Original publisher: [Read original article]() Author: Nico Martin Published: 2026-04-23T00:00:00Z Content type: tutorial Language: en Sources: [Hugging Face - Blog]() Topics: [Chrome extension](), [transformers.js](), [gemma4](), [Artificial Intelligence](), [Local AI](), [Open Source](), [Code]() Tags: [artificial-intelligence](), [chrome](), [chrome-extension](), [code](), [extension](), [github](), [guide](), [inference](), [javascript](), [local](), [local-ai](), [manifest](), [on-device](), [open](), [open-source](), [transformers](), [transformers-js](), [ui](), [web]() ### AI overview A developer guide to building a local AI Chrome extension with Transformers.js under Manifest V3. It explains an architecture with a background service worker hosting models, a side-panel chat interface, and a content script for page extraction and highlighting, using the Gemma 4 Browser Assistant as a reference. ### Source excerpt We're on a journey to advance and democratize artificial intelligence through open source and open science. ## Training and Finetuning Multimodal Embedding & Reranker Models with Sentence Transformers DevFeed: [Training and Finetuning Multimodal Embedding & Reranker Models with Sentence Transformers]() Original publisher: [Read original article]() Author: Tom Aarsen Published: 2026-04-16T00:00:00Z Content type: tutorial Language: en Sources: [Hugging Face - Blog]() Topics: [multimodal](), [sentence-transformers](), [datasets](), [Machine learning](), [Artificial Intelligence](), [Open Source]() Tags: [community](), [data](), [datasets](), [embedding](), [evaluation](), [models](), [multimodal](), [nlp](), [open-source](), [optimization](), [performance](), [retrieval](), [sentence-transformers](), [train](), [training](), [transformers](), [vision]() ### AI overview A practical guide to fine-tuning multimodal Sentence Transformer embedding and reranker models for visual document retrieval. It explains the training pipeline and shows how domain-specific fine-tuning improved retrieval performance from 0.888 to 0.947 in the example evaluation. ### Source excerpt We're on a journey to advance and democratize artificial intelligence through open source and open science. ## Multimodal Embedding & Reranker Models with Sentence Transformers DevFeed: [Multimodal Embedding & Reranker Models with Sentence Transformers]() Original publisher: [Read original article]() Author: Tom Aarsen Published: 2026-04-09T00:00:00Z Content type: article Language: en Sources: [Hugging Face - Blog]() Topics: [sentence-transformers](), [Embeddings](), [multimodal](), [Retrieval Augmented Generation (RAG)](), [vlm](), [GPU](), [Cloud]() Tags: [cloud](), [community](), [embedding](), [gpu](), [multimodal](), [nlp](), [open-source](), [rag](), [retrieval](), [search](), [sentence-transformers](), [transformers](), [vision]() ### AI overview This article explains how multimodal embedding and reranker models in Sentence Transformers map text, images, audio, and video into shared spaces or score cross-modal pairs. It covers visual document retrieval, cross-modal search, multimodal RAG, hardware requirements, model loading, and similarity computation. ### Source excerpt We're on a journey to advance and democratize artificial intelligence through open source and open science. [Next page]()