This guide explains how to quickly launch an open‑source large language model from ModelScope on Alibaba Cloud Function Compute, covering the required cloud services, step‑by‑step deployment, reserved‑instance configuration, and how to invoke the model via the provided domain.
This guide explains how to quickly select a ModelScope open‑source LLM, deploy it to Alibaba Cloud Function Compute using the SwingDeploy one‑click feature, enable reserved idle billing, and evaluate the cost savings compared with traditional GPU provisioning.
Alibaba Cloud's PAI team unveiled Llumnix, a dynamic scheduling framework for large language model serving that dramatically reduces tail latency, speeds high‑priority requests, and cuts costs, earning acceptance at OSDI 2024 and now open‑sourced on GitHub.
This article compiles a comprehensive, up‑to‑date inventory of open‑source large language models from Chinese and international organizations, detailing each model’s architecture, parameter count, multilingual capabilities, deployment requirements, and associated tools, offering a valuable reference for AI researchers and developers.
LMCache separates the KV cache from a vLLM instance into a shared service, dramatically cutting first‑token latency for repeated text, enabling multiple GPU instances to reuse cached vectors, improving hardware utilization, and supporting use cases such as long‑document QA, multi‑GPU load balancing, and prompt‑engineering, with a quick Docker‑based demo.
This guide explains how to deploy a production‑ready DeepSeek‑R1 inference service on Alibaba Cloud ACK using KServe, covering model preparation, storage configuration, service deployment, observability, autoscaling, model acceleration, gray‑release and GPU‑shared inference.
Token‑level Two‑Chunk Overlap replaces traditional batch‑level Two‑Batch Overlap, dynamically splitting sequences into balanced token chunks, enabling near‑equal compute and communication times, improving GPU utilization and achieving up to 30% throughput gains in heterogeneous request workloads, with zero accuracy loss.
FlashAttention‑2 is an IO‑aware exact attention algorithm that cuts GPU HBM traffic through tiling and recomputation, optimizes non‑matmul FLOPs, expands sequence‑parallelism and warp‑level work distribution, delivering up to 2× speedup over FlashAttention, near‑GEMM efficiency, and enabling longer‑context Transformer training and inference for AIGC with fastunet and negligible accuracy loss.
The vLLM 0.17.0 release brings FlashAttention 4 integration, a mature Model Runner V2, complete Qwen 3.5 series support, a one‑click performance‑mode flag, Anthropic API compatibility, advanced weight‑offloading, broader hardware support beyond NVIDIA, ASR model integration, and detailed upgrade and installation guidance.
This guide explains the five common sources of GPU memory consumption in large‑model inference services, provides a step‑by‑step diagnosis workflow—from static usage and KV‑Cache analysis to concurrency and K8s scheduling—offers concrete command‑line checks, scripts, configuration examples, and actionable remediation and monitoring recommendations.
Unsloth offers a unified web UI that accelerates fine‑tuning by up to 2×, cuts VRAM usage by 70% (80% for RL), supports hundreds of open‑source models, and provides simple installation steps for rapid local AI experimentation.
This article details how to deploy the 235‑billion‑parameter Qwen3‑235B model using PD‑separation and MoE techniques, explains the associated challenges, and demonstrates a production‑grade solution built on the high‑performance SGLang inference engine and the RoleBasedGroup (RBG) orchestration framework, complete with benchmark results and best‑practice YAML examples.
This article details the cost and speed challenges of embedding vectors in large‑scale log scenarios, analyzes inference framework choices, describes GPU utilization, priority queuing, and pipeline redesigns, and reports a 16‑fold throughput increase and dramatically lower per‑request costs.
This article derives DeepSeek V3's training Model FLOPs Utilization (MFU) using publicly available data, showing an MFU of roughly 37%—about a 60% improvement over V2—and provides detailed formulas, parameter settings, and a reproducible Python script.
HighService, Alibaba’s Pythonic AI service framework, accelerates large‑model inference and maximizes GPU utilization by separating CPU‑GPU processes, offering out‑of‑the‑box quantization, parallelism and caching, and dynamically reallocating idle GPUs across clusters through a master‑worker scheduler to keep online latency low while boosting offline throughput for diffusion and LLM workloads.
This article explains how to quickly deploy the full‑parameter DeepSeek‑R1 671B model in a multi‑node GPU‑enabled Kubernetes cluster on Alibaba Cloud ACK, covering prerequisites, model parallelism, vLLM‑Ray distributed deployment, service verification, and integration with Dify to build a private AI Q&A assistant.
This article explains how to transform ComfyUI into a serverless API using Alibaba Cloud Function Compute, detailing the challenges of GPU resource costs, high concurrency, and usability, while providing a step‑by‑step guide, code examples, and best‑practice recommendations for building scalable AI drawing applications.
The article dissects DeepSeek‑V4’s local deployment using vLLM, explaining the steep hardware requirements, the complex heterogeneous KV‑cache architecture, and the aggressive kernel‑fusion and multi‑stream optimizations that together make high‑context inference both memory‑intensive and engineering‑heavy.
This article explains how a hierarchical sparse‑attention framework redesigns KVCache storage across GPU, CPU, and remote memory, eliminates bandwidth and capacity bottlenecks, and enables efficient inference for 128K‑token and larger contexts with dramatically reduced GPU memory usage and higher throughput.
The article summarizes Zhihu's machine‑learning platform lead Wang Xin's presentation on the ZhiLight large‑model inference framework, covering model execution mechanisms, GPU workload analysis, pipeline and tensor parallelism, GPU architecture evolution, open‑source engine comparisons, ZhiLight's compute‑communication overlap and quantization optimizations, benchmark results, supported models, and future directions.
