#tokenizer
175 related articles
·7 min
Kontext: One-Click AI Conversation Context Migration, Breaking Down Multi-Model Switching Barriers
Kontext is a context migration tool for multi-AI users, enabling one-click transfer of full conversation history between ChatGPT, Claude, and Gemini to solve the context silo problem.
Read more →
·5 min
Claude Sonnet 5 Integrated with Devin: Enhanced Coding Performance with 30% Lower Quota Consumption
Anthropic's Claude Sonnet 5 launches on Devin Desktop and CLI, delivering frontier-level coding performance while reducing quota consumption by ~30% compared to the previous generation.
Read more →
The Complete Guide to Local LLM Deploy…
·2 min
The Complete Guide to Local LLM Deployment: Tools, Hardware, and Model Selection
A complete guide to locally deploying open-source LLMs: covering VRAM requirements, quantization, tools like Ollama and LM Studio, and model selection tips for Llama, Qwen, and more.
Read more →
·5 min
Introduction to Prompt Engineering: A Complete Guide to Core Concepts, Methodologies, and Limitations
Prompt engineering is a core skill in the AI era. This article breaks down the essential differences between prompts and prompt engineering, the six-step workflow, four evaluation criteria, and key limitations like context limits and hallucination.
Read more →
·5 min
The Attention Bottleneck in AI Coding: Hands-On Agent Workflows from the Father of Crayfish
OpenClaw author Peter shares his thinking on AI coding: when compute is no longer the bottleneck, developer attention becomes the new ceiling. Learn three practical skills—Agent Transcript, Auto-Review, and the Crapbox sandbox.
Read more →
·7 min
Harness Engineering: Core Methodology and Practical Guide for Enterprise-Level AI Programming
Can AI really replace programmers? This article explains Harness Engineering principles and its three evolutionary stages, revealing real pain points of enterprise AI programming.
Read more →
·6 min
Codex Configuration in Practice: A Complete Guide to Accessing GPT-5.5 at Low Cost from China
A detailed guide for Chinese developers on configuring the Codex CLI AI coding tool with GPT-5.5 via API proxies, covering setup steps, efficiency gains, and security risks.
Read more →
·6 min
5 Critical Engineering Pitfalls in Enterprise RAG Knowledge Base Deployment: The Real Barriers from Demo to Production
Why do enterprise RAG knowledge bases dazzle in demos but fail in production? This article dissects five critical engineering pitfalls with real-world case studies from million-doc platforms and ops agents.
Read more →
·6 min
14x Speedup: How Manticore Rebuilt the ONNX Embedding Inference Path
Manticore Search restructured its ONNX inference path to achieve 14x faster text embeddings. Deep dive into batching, session reuse, zero-copy memory, and thread tuning for vector search systems.
Read more →
·6 min
Step-by-Step Guide: Power Codex with DeepSeek for Users Without ChatGPT Access
No ChatGPT account? No problem! Learn how to power Codex with DeepSeek API using the Codex++ management tool. Five easy steps, starting at just $1.40.
Read more →
·6 min
A 284B Model Running on a MacBook: The Real Breakthrough and Barriers of Local DeepSeek Inference
Redis creator runs 284B-parameter DeepSeek model on a MacBook Pro at 26 tokens/sec using a pure C engine, asymmetric quantization, and MoE architecture.
Read more →
·3 min
Is Claude Sonnet 5 Really Cheap? A Cost Breakdown by Task
Claude Sonnet 5 markets itself on agentic capabilities and low price, but real costs are far more complex. We break down token explosion, tokenizer inflation, and Opus 4.8 comparisons to reveal the true cost-performance picture.
Read more →
·3 min
Claude Sonnet 5 Hands-On Review: The Hidden Token Trap That Makes 'Low Cost' an Illusion
Claude Sonnet 5 promises near-Opus 4.8 performance at lower cost, but hands-on tests reveal a critical trap: a new tokenizer inflates token consumption, making real costs far higher than expected.
Read more →
·3 min
Needle: Fine-Tuning a 26M Ultralight Tool-Calling Model Locally — Full Walkthrough
Needle is a 26M-parameter tool-calling model. Learn how to replace Gemini with Ollama for local training data generation and fine-tune on a single GPU, achieving 96.7% F1 — ideal for edge AI deployment.
Read more →
·1 min
Content Mismatch Detected — Article Cannot Be Generated
Source material provided (song lyrics) is unrelated to the target technical topic. Unable to generate an AI Agent / GPT architecture article. Please resubmit with valid technical content.
Read more →
·7 min
Building an AI Agent from Scratch: Implementing Core Conversational Agent Features in Python
Learn AI Agent development from scratch. This tutorial covers LLMs and prompts, then builds a conversational agent in Python using the DeepSeek API with multi-turn dialogue and system prompts.
Read more →
Has AI Coding Assistant Fable Been Ner…
·4 min
Has AI Coding Assistant Fable Been Nerfed? Developer's Routing Anomaly Report Sparks Debate
A developer questioned whether AI coding assistant Fable was nerfed, finding 4–10x more tokens routed to Opus with Fable doing ~20% of the work. A deep dive into multi-model routing, transparency, and AI trust.
Read more →
·7 min
Cost Optimization for AI Agent Fleets: Why an Expensive Commander + Cheap Workers Is More Cost-Effective
Deep analysis of multi-agent system cost optimization: why the 'expensive commander + cheap workers' combination outperforms all-frontier fleets, covering decision-intent cost logic and Sonnet 5 tokenizer traps.
Read more →
Is Claude Sonnet 5 Worth Upgrading To?…
·3 min
Is Claude Sonnet 5 Worth Upgrading To? A Developer's Guide to Performance vs. Cost
Claude Sonnet 5 approaches Opus-level performance, but a new tokenizer increases token usage by ~30%. This guide helps developers rationally evaluate the upgrade.
Read more →
·8 min
Free Top-Tier AI Models on Aggregator Platforms? A Sober Analysis and Pitfall Avoidance Guide
In-depth analysis of AI aggregator platforms claiming free access to GPT, DeepSeek, and Gemini. Reveals hidden data risks, business logic, and recommends legitimate alternatives like OpenRouter and Poe.
Read more →