Skip to main content
株式会社オブライト

Column

Useful articles about SEO, Web Development, and IT

771 articles

Business DX2026-07-09
How to Use Chamber of Commerce and Municipal Digitalization Support Programs
A neutral overview of chambers of commerce, Yorozu support centers, and municipal windows that support SME digitalization — how they differ and how to use them.
地方創生中小企業DX支援制度
Business DX2026-07-09
The Minimum IT Toolkit for Small Companies — How to Choose a Standard Set
A neutral, category-by-category look at the minimum IT toolkit small companies with just a few employees should set up first, without endorsing specific products.
中小企業DXITツール
Business DX2026-07-09
Shinagawa Ward SME Support & DX Programs Guide — From Manufacturing to IT
A guide to the types of SME support in Shinagawa Ward — consultation, expert dispatch, subsidies, startup support — how to find them, and a general approach to digitalization for local businesses.
中小企業DX品川区東京6区
Business DX2026-07-09
Minato Ward SME Support & DX Programs Guide — Options for Small Central-Tokyo Businesses
A guide to the types of SME support in Minato Ward — consultation, expert dispatch, subsidies, loan referrals — how to find them, and a general approach to digitalization for small central-Tokyo businesses.
中小企業DX港区東京6区
Business DX2026-07-09
Shibuya SMB & Startup Support Guide: How to Find IT and Founding Assistance Programs
A neutral guide for SMBs and startups in Shibuya on the types of business, founding, and IT support programs available, and how to find them.
中小企業DX渋谷区東京6区
Business DX2026-07-09
Setagaya SMB Support & DX Guide: Finding Assistance for Shopping Districts and Small Businesses
A neutral guide for Setagaya's shopping district businesses and small operators on the types of SMB support available and how to find them, including digitalization steps.
中小企業DX世田谷区東京6区
Business DX2026-07-09
Meguro Ward SME Support and DX Program Guide
A neutral guide to the types of support programs available to SMEs in Meguro Ward, how to find them, and general steps for digitalization. Always check the ward's official sources for the latest details.
中小企業DX目黒区東京6区
Business DX2026-07-09
Ota Ward's Small Factories and DX: Support Programs and Practical Steps for a Manufacturing Town
A neutral guide to how Ota Ward's small factories and manufacturing SMEs can find support programs, plus practical steps for starting DX without overreaching.
中小企業DX大田区製造業+1
Software Development2026-07-09
Crit.md Deep Dive — A Local-First Review Tool That Gives AI-Coding-Agent Output the "Pull-Request Review" Experience, With First-Class Support for Claude Code / Cursor / Copilot / Codex / Gemini / Qwen Four Review Modes (Plans & Docs, Code Diffs, Live Apps, Static HTML), Per-Line Comments Persistent Across Iteration Rounds
Crit.md is a local-first review tool that gives AI-coding-agent output (Claude Code / Cursor / GitHub Copilot / Codex / OpenCode / Gemini / Qwen / ...) a "pull-request review" experience in the browser. Four review modes in one UI: Plans & docs (Markdown rendering), Code diffs (syntax highlighting), Live apps (comment directly on the DOM), and Static HTML preview. Per-line comments ("click line 47, type feedback, the agent fixes it"), comments persist across iteration rounds, no account required / binds to 127.0.0.1 / zero telemetry, and a file-based protocol (structured review files exchanged with agents — no vendor APIs, no lock-in). Target users: engineers, staff developers, engineering managers. Reduces time spent re-reading terminal diffs, and keeps humans in the loop instead of delegating all review to agents. Sits opposite the "fully autonomous" agents in our Claude Cowork and Devin coverage — a HITL review substrate for AI output.
Crit.mdAI Coding AgentsCode Review+4
Mobile Development2026-07-09
React Native ExecuTorch's useLLM Hook Deep Dive — Run Qwen / Llama 3.2 / Hammer 2.1 / Phi 4 Mini / SmolLM 2 / LFM2.5 / Gemma 4 On-Device in React Native, With Tool Calling, Vision / Audio, and Structured Output From Software Mansion, Two Modes (Managed / Functional), and Zod-Backed Schema Validation
Software Mansion's React Native ExecuTorch shipped a useLLM hook that gives React Native apps native on-device LLM integration. Supported models: quantized Qwen (2.5 / 3 / 3.5), Llama 3.2, Hammer 2.1, Phi 4 Mini, SmolLM 2, LFM2.5 (vision), and Gemma 4 (vision + audio). Two operating modes: Functional/Stateless (developers manage conversation history via generate() + response; tool calling and chat config don't apply) and Managed/Stateful (sendMessage() maintains conversation state, parses tool calls, and runs callbacks automatically). Key features: token batching (groups tokens before re-render), tool calling (model invokes external functions via tool schemas with automatic parsing and callbacks), vision-language / audio multimodal inputs, generation control (temperature / top-p / repetition penalty / mid-stream interruption), and JSONSchema- or Zod-backed structured output. Use cases: on-device chatbots without server dependencies, privacy-first conversation UIs, in-app function calling (calendar events, flashlight, etc.), multimodal features (image analysis, audio transcription). Positioning: implementation-level evidence that cutting-edge open-weight LLMs like Gemma 4's encoder-free 12B and Qwen 3.6-35B natively fit into iOS / Android apps — a major deliverable from the React Native ecosystem.
React NativeExecuTorchOn-Device LLM+5
AI2026-07-09
Nous Portal Deep Dive — One OAuth Login for 300+ Frontier AI Models (Claude / GPT / Gemini / DeepSeek / Qwen / Kimi / GLM / MiniMax / Grok) Plus a Tool Gateway (Web Search, Image Generation, TTS, Browser Automation), the Official Backend for Hermes Agent Launched 2026-04-27, Free $0 / Plus $20 With $22 in Rollover-Enabled Credits
Nous Research's Nous Portal launched on 2026-04-27 as an AI-gateway subscription. A single OAuth login unlocks 300+ frontier, agentic-oriented models (Claude, GPT, Gemini, DeepSeek, Qwen, Kimi, GLM, MiniMax, Grok, and more), plus a Tool Gateway (web search, image generation, TTS, browser automation) and Nous Chat. Designed as the official backend for Nous Research's Hermes Agent, it removes per-vendor API-key juggling. Pricing: Free $0/month ($0.10 credit) and Plus $20/month (300+ models, hosted tools, $22 worth of monthly credits with rollover). API access with API Docs enables programmatic integration through an OpenAI-compatible API on a monthly-credit model. Strategic significance: the AI-side equivalent of the integration trend from our "just deploy everything to Cloudflare" column — a "run everything through Nous Portal" approach that avoids per-vendor lock-in and gives fast access to the newest models, with rollover credits making cost management manageable. Caveats: dependency on Nous Portal itself, coupling with the Hermes Agent ecosystem, and data-protection terms remain those of each underlying model vendor at the enterprise contract layer.
Nous ResearchNous PortalHermes Agent+3
AI2026-07-08
Gemma 4 Technical Report Deep Dive — Google DeepMind's Open-Weight, Natively Multimodal 2.3B–31B LLMs with an Encoder-Free 12B Unified Design and Built-In Reasoning Mode arXiv:2607.02770, Published 2026-07-02, 300+ Authors, Both Dense and MoE Variants
Google DeepMind's Gemma Team released the Gemma 4 Technical Report as arXiv:2607.02770 on 2026-07-02. The paper introduces 2.3B / 12B / 31B parameter models, both Dense and MoE variants, natively multimodal (text / image / audio), a 12B encoder-free unified design (raw audio and image patches processed directly without separate encoders), a built-in reasoning (thinking) mode, improved vision / audio encoders, architectural refinements for inference speed, memory efficiency, and long context, and competitive performance against larger open models on STEM, multimodal, and long-context benchmarks. Over 300 authors contributed. Open weights allow commercial use, distributed via Hugging Face and Ollama. Sits alongside Qwen 3.6-35B-A3B and the Local LLM June 2026 update as a new chapter at the open-weights frontier. A milestone in Google's open-weights strategy; the encoder-free unified design departs from Qwen / Llama / DeepSeek multimodality (separate vision encoder + projection). The reasoning mode mirrors the extended-thinking modes of Anthropic and OpenAI closed models — the open ecosystem catching up. Caveats: commercial-license fine print, potential systemic-risk classification (EU AI Act's 10^25 FLOPs threshold), and heavy Google Cloud Vertex AI integration bias.
Gemma 4Google DeepMindOpen Weight+3
1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465