Skip to main content
株式会社オブライト

Column

Useful articles about SEO, Web Development, and IT

Browse by calendar (flip like a book)

895 articles

Business DX2026-09-16
Employee Safety Confirmation System Costs for SMBs (2026)
Safety confirmation system costs for SMBs: SaaS runs 100-500 yen/user monthly, packages 300K-2M yen, custom 3M+ yen. Covers features and selection tips.
中小企業DXITリスク対策システム発注+1
Software Development2026-09-15
What Is dbt Charts? YAML Dashboards for AI Agents, Explained
dbt Labs open-sourced dbt Charts in Sept 2026: YAML for SQL dashboards, Apache-2.0. Covers install, init-validate-serve-render flow, AI agent skills, and MCP.
開発ツールオープンソースAIエージェント+3
Business DX2026-09-15
HR Performance Evaluation System Costs for SMBs (2026)
HR evaluation system costs run from a few hundred yen per employee for SaaS to millions for custom builds, covering MBO, OKR, and 360-degree fit for SMBs.
中小企業DXシステム発注開発費用+1
Business DX2026-09-14
Equipment Maintenance System (CMMS) Cost Guide for SMBs (2026)
CMMS costs range from a few hundred yen per asset monthly for cloud tools to millions of yen for custom IoT builds, with cost drivers and a manager checklist.
中小企業DXシステム発注開発費用+2
Web Development2026-09-14
Three.js r186: Native Gaussian Splatting, Setup & Limits
Three.js r186 adds GaussianSplat, a built-in mesh rendered via WebGPURenderer with GPU depth sorting. Covers setup, minimal code, and Spark 2.0 differences.
Three.jsWebGPUTSL+3
AI2026-09-14
What Is colibri? The Pure-C Engine That Runs a 744B MoE Model on 25GB RAM
A guide to JustVugg/colibri, a pure-C engine that hit GitHub Trending by running the 744B GLM-5.2 MoE model on consumer hardware via on-demand expert streaming from disk. Covers how it works, supported models, build/usage, hardware requirements, and how it differs from llama.cpp, Ollama, Edge0, and turbo-fieldfare.
colibriローカルLLMMoE+4
Software Development2026-09-14
What Is Context Mode? The MCP Server That Cuts AI Agent Tool Output by 98%
A guide to mksglu/context-mode, which passed 22,000 GitHub stars with +2,102 in a single week. Covers how it sandboxes tool output to cut context window consumption, how to install it on Claude Code, and how it compares to OpenViking, ECC, and the built-in /compact command.
MCPClaude CodeAIエージェント+2
Software Development2026-09-14
What Is CUDA for AMD on Windows? Running CUDA Apps on AMD GPUs via ZLUDA + HIP/ROCm
A guide to Speedstu/CUDA-for-AMD-Windows, which hit 135 points on Hacker News. Covers how this reproducible ZLUDA + AMD HIP/ROCm stack runs CUDA-targeted Windows apps on AMD GPUs, validated hardware, installation, what works and what doesn't, and how it compares to plain ZLUDA, ROCm on Linux, DirectML, and Vulkan-based llama.cpp.
NVIDIAAMDローカルAI+2
AI2026-09-14
What Is Edge0-35B-A3B? How SSD Streaming Runs a 35B MoE Model in Under 3GB of RAM
Released Sep 2026, Edge0-35B-A3B-preview 4-bit-quantizes Qwen3.5-MoE 35B-A3B and streams experts off SSD on demand, running at under 3GiB of active memory and roughly 15-18 tok/s on MLX. Here's how it works, what hardware you need, how to run it, and how it compares to llama.cpp mmap offload and Colibri.
RequirementsVRAMローカルLLM+4
AI2026-09-14
What Is GPT Image 2.5? Flare vs Sunburst, API Pricing, and Per-Image Cost
A guide to GPT Image 2.5 (ChatGPT Images 2.5), released by OpenAI on September 8, 2026. Covers the difference between the fast Flare model and the precision-focused Sunburst model, official API token pricing, estimated per-image cost, and how it compares to Gemini 3 Pro Image and Midjourney.
OpenAIAPI料金画像生成+1
Software Development2026-09-14
What Are i-have-adhd and Ponytail? Inside GitHub's Two Trending Coding-Agent Skills (Sep 2026)
A guide to two Agent Skills for coding agents that surged on GitHub's weekly trending charts: i-have-adhd (+16,740 stars/week), which stops the agent from burying its answer and leads with the next action, and ponytail (+8,444 stars/week), which makes the agent think like the laziest senior developer in the room and avoid writing unnecessary code. Covers what's inside each SKILL.md, how to install them in Claude Code and Codex, and how they differ from grill-me, superpowers, and humanizer.
Agent SkillsClaude Codeコーディングエージェント+2
AI2026-09-14
MiniCPM5-2B Requirements: RAM, VRAM and GPU for Phones, Raspberry Pi and Laptops (2026)
As of Sep 7, 2026, OpenBMB's MiniCPM5-2B (2.52B dense, 131K context, Apache-2.0) needs roughly 1.5GB to 5GB of RAM/VRAM depending on quantization. Here is a GGUF size table, how to run it with Ollama and llama.cpp, and a comparison with Qwen3.5-4B.
MiniCPMRequirementsVRAM+2
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475