Author

Baidu Geek Talk

511

Articles

Likes

879

Views

Comments

Latest from Baidu Geek Talk

100 recent articles max

Baidu Geek Talk

Apr 2, 2025 · Artificial Intelligence

DeepSeek-VL2 Multimodal Model: Architecture, Training, and Code Walkthrough

DeepSeek‑VL2 is a state‑of‑the‑art multimodal model built on a Mixture‑of‑Experts architecture that combines a SigLIP‑L vision encoder with dynamic tiling, a two‑layer VL adaptor, and a DeepSeek‑MoE language model using Multi‑head Latent Attention, trained in three stages on diverse visual‑language and text data, and achieving strong results on benchmarks such as DocVQA and TextVQA, with full implementation and inference code available in PaddleMIX.

DeepSeek-VL2InferenceMixture of Experts

0 likes · 36 min read

DeepSeek-VL2 Multimodal Model: Architecture, Training, and Code Walkthrough

Baidu Geek Talk

Mar 24, 2025 · Big Data

How Turing Data Finder Transforms Growth Analysis with a Unified Data Platform

The article provides a detailed technical overview of the Turing Data Finder (TDF) platform, describing its background, core components, data schema, ingestion workflow, and a suite of growth‑analysis features such as event, retention, funnel, path, component, distribution, and attribution analysis, while also outlining performance‑optimisation techniques and future development directions.

SQL OptimizationTuring Data Finderbig data

0 likes · 17 min read

How Turing Data Finder Transforms Growth Analysis with a Unified Data Platform

Baidu Geek Talk

Mar 19, 2025 · Artificial Intelligence

Inside Baidu’s New Wenxin 4.5 & X1: Multimodal Breakthroughs and Tool‑Enabled AI

Baidu officially launched the Wenxin 4.5 and X1 large language models, showcasing native multimodal foundations, advanced attention masks, heterogeneous expert extensions, and tool‑calling capabilities, while offering low‑cost API access on the Qianfan platform and outlining the underlying technical innovations that drive their performance gains.

AI PlatformBaiduTool Calling

0 likes · 8 min read

Inside Baidu’s New Wenxin 4.5 & X1: Multimodal Breakthroughs and Tool‑Enabled AI

Baidu Geek Talk

Mar 17, 2025 · Industry Insights

From Manual Restarts to Automated Fault Tolerance: The Evolution of AI Training Stability

This article traces the decade‑long evolution of AI training stability—from early small‑model manual operations to large‑scale, multi‑thousand‑GPU clusters—detailing metrics like invalid training time, fault‑tolerance architectures, eBPF‑based hidden‑fault detection, BCCL enhancements, multi‑level restart strategies, and trigger‑based checkpointing that together shrink downtime from minutes to seconds.

AI trainingInfrastructuredistributed-systems

0 likes · 22 min read

From Manual Restarts to Automated Fault Tolerance: The Evolution of AI Training Stability

Baidu Geek Talk

Mar 12, 2025 · Artificial Intelligence

How LLMs Are Revolutionizing Semantic Embeddings: Models, Methods, and Trends

This article reviews how large language models (LLMs) enhance semantic text embeddings by comparing traditional methods with LLM‑based approaches, detailing synthetic data generation, backbone model designs, key model families, experimental results on the MTEB benchmark, and future research challenges.

LLMcontrastive learningmodel comparison

0 likes · 30 min read

How LLMs Are Revolutionizing Semantic Embeddings: Models, Methods, and Trends

Baidu Geek Talk

Mar 10, 2025 · Artificial Intelligence

How Baidu Cloud’s AI+ Strategy Powers State‑Owned Enterprises with DeepSeek One‑Box Solutions

The article examines Baidu Cloud’s integration of DeepSeek large‑model hardware, detailing the Baige and Qianfan one‑box systems, their technical specs, deployment speed, and how they enable state‑owned enterprises across energy, manufacturing, and logistics to accelerate AI‑driven digital transformation.

AIBaidu CloudDeepSeek

0 likes · 6 min read

How Baidu Cloud’s AI+ Strategy Powers State‑Owned Enterprises with DeepSeek One‑Box Solutions

Baidu Geek Talk

Mar 5, 2025 · Cloud Computing

Inside GPU Cloud Servers: Architecture, Interconnects, and Performance Secrets

This article provides a comprehensive technical overview of GPU cloud server design, covering data‑processing pipelines, hardware topology, NUMA considerations, PCIe and proprietary interconnects, multi‑GPU communication strategies, virtualization approaches (BCC and BBC), DPU acceleration, and future trends for scaling up and out.

GPUInterconnectPerformance optimization

0 likes · 27 min read

Inside GPU Cloud Servers: Architecture, Interconnects, and Performance Secrets

Baidu Geek Talk

Mar 3, 2025 · Big Data

Real-Time Anti-Cheat Streaming System Based on Flink: Architecture, Challenges, and Solutions

The article details a Flink‑based real‑time anti‑cheat streaming architecture that combines tumbling, sliding and session windows with early triggers, batch state updates cached in memory, coarse‑grained key reduction, and YAML‑driven strategy configuration to deliver millisecond‑level detection, seamless integration with ClickHouse, Hive, Redis and message queues, and self‑service analytics, achieving high throughput, low latency, and robust stability for large‑scale risk control.

Configuration ManagementFlinkPerformance optimization

0 likes · 25 min read

Real-Time Anti-Cheat Streaming System Based on Flink: Architecture, Challenges, and Solutions

Baidu Geek Talk

Feb 26, 2025 · Artificial Intelligence

Baidu Cloud's DeepSeek Integrated Solutions for Enterprise AI Transformation

Baidu Cloud’s new DeepSeek integrated machine suite—Bai Ge, Qian Fan and Yi Jian—combines domestic Kunlun P800 chips with one‑click private deployment of full DeepSeek models, delivering high cost‑performance, security compliance and end‑to‑end AI capabilities for diverse industry digital transformation.

AI transformationDeepSeek ModelEnterprise Solutions

0 likes · 8 min read

Baidu Cloud's DeepSeek Integrated Solutions for Enterprise AI Transformation

Baidu Geek Talk

Feb 24, 2025 · Artificial Intelligence

Using a Graph Engine to Drive Workflow for Intelligent Agents

By leveraging mature graph‑engine technology, the article shows how visual, low‑code workflow orchestration can give intelligent LLM‑based agents fine‑grained path control, reusable functions, hierarchical sub‑flows, and robust error handling, turning complex business tasks into modular, scalable processes adopted by hundreds of thousands of developers.

AI agentsLLMgraph engine

0 likes · 18 min read

Using a Graph Engine to Drive Workflow for Intelligent Agents