#LLM Token optimization
1217 related articles
·3 min
Reranking: The Underrated Secret Weapon for Boosting RAG Accuracy
Explore why reranking is the key to boosting RAG retrieval accuracy. Learn how cross-encoders overcome vector search limitations and how to implement two-stage retrieval architectures.
Read more →
·3 min
Maple-Preview: 20B Ternary MoE Model Achieves 120 tok/s On-Device Inference on iPhone
Maple-Preview achieves 120 tok/s inference of a 20B ternary MoE model on iPhone. We analyze ternary quantization, MoE sparse activation, and on-device inference challenges.
Read more →
·3 min
Cloudflare Uses AI to Standardize Engineering Practices: From Standards Drift to Automated Enforcement
Deep dive into how Cloudflare uses LLMs to auto-enforce engineering standards, solving standards drift in large teams. Explores AI code review in CI/CD pipelines, challenges, and implications.
Read more →
·3 min
The 5-Layer AI Tech Stack Explained: A Complete Industry Chain from Energy to Applications
Deep dive into the 5-layer AI tech stack: Energy, Chips, Infrastructure, Models, and Applications. Understand the key players, competitive landscape, and value distribution logic across the AI industry chain.
Read more →
·3 min
Quantprobe: An Open-Source Tool for Running 30B LLMs on 6GB GPUs
Quantprobe is an open-source memory optimization framework that enables 30B LLMs to run at 22 tokens/s on 6GB GPUs through per-layer quantization and intelligent CPU/GPU splitting.
Read more →
·2 min
The Real Experience of AI Programming: Genuine Productivity Boost or False Prosperity?
An in-depth analysis of AI programming tools' real value and limitations: from boilerplate acceleration to hallucination issues, from efficiency illusions to complex system failures—a sober assessment from a frontline developer's perspective.
Read more →
·3 min
Harness Engineering: The Key Technique for Unlocking Hidden AI Agent Capabilities
Explore how harness engineering dramatically improves AI Agent performance. From the Codex case study, learn how tool orchestration, context management, and execution environments become the core competitive battleground.
Read more →
·2 min
Qwen3.8-Max Preview Continues Iterating with Major Improvements in Frontend Development Capabilities
Alibaba's Qwen3.8-Max-Preview iterates daily with significant frontend development improvements. The team uses an open preview strategy to collect community feedback, promising open-weight release.
Read more →
·3 min
Comparing China's Four Major AI Labs: Strategic Analysis of Qwen, DeepSeek, Moonshot, and Ling
An insider's analysis of China's four AI labs — Qwen, DeepSeek, Moonshot, and Ling — revealing their distinct strategic bets on distribution, architecture, long-termism, and serving cost.
Read more →
·2 min
Cursor Hidden Markup Controversy: Cached Tokens Charged 12.5x Surcharge
A Cursor enterprise power user exposes hidden markups in its Luna pricing: cached tokens charged a 12.5x surcharge. Detailed breakdown of the Cursor Tax billing logic and its real impact on users.
Read more →
·4 min
Running an 80B Large Model on Mac with Only 4.3GB of Memory: Extreme Compression Techniques Explained
Deep dive into how an 80B-parameter LLM runs on Mac with only 4.3GB memory, covering ultra-low-bit quantization, sparsity, memory mapping, and implications for privacy and edge AI.
Read more →
·3 min
Xberg v1 Open-Source Document Extraction Engine: CPU-Only Local Processing Supporting 101 Formats
Xberg v1 is an MIT-licensed open-source local document extraction engine. CPU-only, supporting 101 formats with built-in SPLADE and ColBERT retrieval, Rust-powered for RAG and ML pipelines.
Read more →
·3 min
LiveKit Agents: A Comprehensive Guide to the Open-Source Framework for Building Real-Time Voice AI Agents
Deep dive into the LiveKit Agents open-source framework for building real-time voice AI agents using STT, LLM, and TTS modules with production-ready deployment capabilities.
Read more →
·3 min
MCP Session Analytics: Building Production-Grade Observability and Evaluation for AI Agents
How to build product analytics and evaluation capabilities for AI Agents at the MCP protocol layer, covering session-level tracing, tool call observability, and quality Evals.
Read more →
·2 min
MicroCodex: A C++ Programming Agent Under 1MB — A Fresh Take on Minimalist AI Tools
MicroCodex re-implements OpenAI's Codex programming agent in C++ with a binary under 1MB. This article analyzes its engineering design, minimalist philosophy, and use cases in CI/CD and embedded systems.
Read more →
·3 min
AI Tool Mashups in Practice: Combining Model Layering with Intelligent CLI
Explore AI development tool mashups: model layering with DeepSeek Flash, flagship model selection, Antigravity CLI, and practical strategies for model routing and tool composition.
Read more →
·3 min
Bolcho AI: A Voice Agent Platform Built for India's Multilingual Market
Bolcho AI is a voice AI platform for India's market, supporting Hindi, Tamil and more local languages with ultra-low latency, telephony integration, and flexible BYO model architecture for enterprise AI agents.
Read more →
·2 min
The 'Laziness' Controversy Around GPT-5.6 Sol: 70% of Runtime Spent Executing Sleep Commands
Developers found GPT-5.6 Sol spends ~70% of runtime on sleep commands, sparking debate about balancing model caution vs. efficiency in the AI agent era.
Read more →
·2 min
Claude Opus 5 Arrives on Devin: Fable-Level Performance at Half the Cost
Devin integrates Claude Opus 5, achieving near Fable-level performance on FrontierCode 1.1 at half the cost. The model excels at difficult debugging and root-cause analysis across Desktop, CLI, and Cloud.
Read more →
·2 min
Tempest Open-Source Tool: Reduces Token Consumption by 64% for Parallel AI Coding Agents
Tempest is an open-source developer tool that reduces token consumption by up to 64% for parallel AI coding agents through shared code understanding and isolated workspaces.
Read more →