#stop_reason
6 related articles
·3 min
Amazon Bedrock Converse API in Practice: A Development Guide to Unified Interfaces and Streaming Chat
A deep dive into Amazon Bedrock's Converse API unified multi-model interface and ConverseStream streaming output, covering message structures, multi-turn conversations, event stream handling, and AWS ecosystem integration.
Read more →
·3 min
Claude Code Core Decoded: Implementing an Agent Loop in 30 Lines of Code
Deep dive into Claude Code's Agent Loop core mechanism, implementing a minimal agent loop in under 30 lines of code. Master AI agent development fundamentals from stop_reason signal-driven design to unified Bash tool entry.
Read more →
·3 min
Unsloth Major Update: DiffusionGemma & Gemma 4 MTP Support, 2x Inference Speed Boost
Unsloth v0.1.464-beta adds DiffusionGemma, Gemma 4 MTP, and MiniMax-M3 support, delivering ~2x inference speed boost, new Hub, RAG Q&A, tensor parallelism, and full CUDA/ROCm/Windows coverage.
Read more →
GPT-5.5 Codex Reasoning-Token Clusteri…
·2 min
GPT-5.5 Codex Reasoning-Token Clustering Explained: Causes of Performance Regression and How to Respond
An in-depth look at reasoning-token clustering in GPT-5.5 Codex — exploring causes like reasoning budget imbalance and model iteration side effects, with prompt engineering strategies for developers.
Read more →
·3 min
Claude Code Local Deployment: A Practical Guide to Connecting Local LLMs
Complete guide to deploying Claude Code locally with Ollama, LM Studio, or vLLM. Covers architecture, protocol translation, hardware requirements, and model selection for zero-cost, private AI coding.
Read more →
Deep Dives·4 min
Agent Loop Core Mechanism Explained: Building a ReAct Loop from Scratch (With Debugging War Stories)
Deep dive into Agent Loop, the core mechanism of AI Agents, based on ReAct architecture. Covers Tool Call ID handling, context management, assistant message feedback, with full Python code and common bug fixes.
Read more →