BestHub
Sign in
Old Zhang's AI Learning
Author

Old Zhang's AI Learning

AI practitioner specializing in large-model evaluation and on-premise deployment, agents, AI programming, Vibe Coding, general AI, and broader tech trends, with daily original technical articles.

141
Articles
0
Likes
3
Views
0
Comments
Recent Articles

Latest from Old Zhang's AI Learning

100 recent articles max
Old Zhang's AI Learning
Old Zhang's AI Learning
Mar 12, 2026 · Artificial Intelligence

Distilling Claude Opus 4.6 into Qwen3.5‑27B: High‑Quality Reasoning on a Single RTX 3090

The article details how Claude Opus 4.6's chain‑of‑thought data were used to distill the 27‑billion‑parameter Qwen3.5‑27B model with Unsloth and LoRA, achieving full‑context inference on a single RTX 3090/4090, while outlining performance numbers, hyper‑parameter tips, benchmark gains and the trade‑offs of losing multimodal abilities.

Claude Opus 4.6GPU inferenceLoRA
0 likes · 7 min read
Distilling Claude Opus 4.6 into Qwen3.5‑27B: High‑Quality Reasoning on a Single RTX 3090
Old Zhang's AI Learning
Old Zhang's AI Learning
Mar 11, 2026 · Artificial Intelligence

Upgrade All Your Claude Skills Now: Harness the New Skill‑Creator Engine

Anthropic’s updated skill‑creator turns Skills into a core, engineering‑focused capability for Claude, offering a systematic workflow—baseline A/B testing, quantitative assertions, visual evaluation, and iterative description optimization—so developers can rebuild, refine, and reliably trigger their Skills for higher productivity.

AI agentsAnthropicAutomation
0 likes · 13 min read
Upgrade All Your Claude Skills Now: Harness the New Skill‑Creator Engine
Old Zhang's AI Learning
Old Zhang's AI Learning
Mar 10, 2026 · Artificial Intelligence

FireRed-OCR 2B: An Open‑Source VLM That Tackles Structural Hallucination

FireRed‑OCR‑2B, an open‑source 2‑billion‑parameter visual‑language model, addresses structural hallucination in document OCR through a geometry‑aware data factory and a three‑stage training pipeline, achieving a 92.94 OmniDocBench v1.5 score and leading end‑to‑end performance while remaining lightweight enough for consumer‑grade GPUs.

FireRed-OCROCROmniDocBench
0 likes · 11 min read
FireRed-OCR 2B: An Open‑Source VLM That Tackles Structural Hallucination
Old Zhang's AI Learning
Old Zhang's AI Learning
Mar 10, 2026 · Artificial Intelligence

Install AutoClaw in One Minute: Quick Setup for a Local AI Assistant

AutoClaw wraps the open‑source OpenClaw client, turning a half‑day installation into three simple steps—download, install, and auto‑configure—while adding seamless Feishu integration, support for GLM‑5 and pony‑alpha‑2 models, built‑in skills, and security recommendations for custom skill creation.

AI AssistantAutoClawFeishu
0 likes · 6 min read
Install AutoClaw in One Minute: Quick Setup for a Local AI Assistant
Old Zhang's AI Learning
Old Zhang's AI Learning
Mar 9, 2026 · Artificial Intelligence

Why You Should Upgrade to OpenClaw v2026.3.8 Today

The v2026.3.8 release of OpenClaw adds official backup commands, improves remote gateway configuration, introduces a configurable Talk silence timeout, refines the TUI and Brave search integration, and fixes numerous stability issues, making the personal AI assistant far more reliable for long‑term use.

AI AssistantBackupOpenClaw
0 likes · 11 min read
Why You Should Upgrade to OpenClaw v2026.3.8 Today
Old Zhang's AI Learning
Old Zhang's AI Learning
Mar 8, 2026 · Artificial Intelligence

Andrew Ng’s New “Context Hub” Adds a Context Layer for Coding Agents

Context Hub, Andrew Ng’s latest open‑source project, introduces a structured, versioned documentation layer for coding agents that tackles API hallucination and memory loss, offers incremental fetching, annotation, and separates local notes from public feedback, making agent‑driven development more reliable and token‑efficient.

AI documentationAnnotationCLI
0 likes · 9 min read
Andrew Ng’s New “Context Hub” Adds a Context Layer for Coding Agents
Old Zhang's AI Learning
Old Zhang's AI Learning
Mar 8, 2026 · Artificial Intelligence

Twinkle – A Lightweight, Fully Chinese Large‑Model Training Framework from ModelScope

Twinkle is a lightweight client‑server training framework open‑sourced by ModelScope that abstracts away Ray clusters, data and model parallelism, offers three run modes (torchrun, Ray, HTTP), multi‑tenant LoRA training, dual back‑ends (Transformers and Megatron), and a serverless Training‑as‑a‑Service gateway for enterprise and individual developers.

LoRAModelScopeMulti‑Tenant
0 likes · 14 min read
Twinkle – A Lightweight, Fully Chinese Large‑Model Training Framework from ModelScope
Old Zhang's AI Learning
Old Zhang's AI Learning
Mar 7, 2026 · Artificial Intelligence

vLLM 0.17.0 Release: Full Qwen 3.5 Support and Anthropic API Compatibility

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.

ASRAnthropic APIFlashAttention
0 likes · 12 min read
vLLM 0.17.0 Release: Full Qwen 3.5 Support and Anthropic API Compatibility
Old Zhang's AI Learning
Old Zhang's AI Learning
Mar 7, 2026 · Artificial Intelligence

5 Key Takeaways After Deep‑Diving the Official Codex Windows Docs

The article redefines Codex from a simple code‑completion tool to a supervised, configurable, parallel, and auditable software‑engineering agent, outlines its evolution, official high‑level usage, why community tutorials fall short, and provides a step‑by‑step guide for deep, production‑grade adoption.

AGENTS.mdAI agentsOpenAI Codex
0 likes · 24 min read
5 Key Takeaways After Deep‑Diving the Official Codex Windows Docs