Skip to main content
株式会社オブライト

Column

Useful articles about SEO, Web Development, and IT

603 articles

Business DX2026-08-03
Workflow & Approval System Cost Guide 2026: SMB Edition
A cost guide to workflow approval systems for SMBs. Compares SaaS fees with custom builds from one to eight million yen, plus cost drivers and choice tips.
システム発注中小企業DX開発費用+1
AI2026-08-03
MiniMax H3 Requirements: VRAM, GPU Sizing & File Sizes (2026 Open-Weight Video+Audio Model)
MiniMax H3, an open video+audio model, released weights Aug 3, 2026. ComfyUI needs ~42.5GB files, ~24GB VRAM (12GB may work). Covers quantization, GPU sizing.
MiniMaxRequirementsVRAM+4
AI2026-08-03
Qwen3.7 Flash Explained: $0.03 per Million Tokens, 1M Context, and What This Closed Vision-Language Model Can (and Can't) Do (2026)
Qwen3.7 Flash: Alibaba's closed vision model, priced from $0.03 per million input tokens, 1M-token context. Tiered pricing, use cases, limits. Updated Aug. 2026.
Qwen 3.7AlibabaAPI料金+3
Business DX2026-08-02
Expense Report System Costs: SaaS vs. Custom Development
A neutral guide to expense report system costs: off-the-shelf SaaS vs. custom development, covering OCR, IC card integration, and storage law compliance.
システム発注中小企業DX開発費用+1
AI2026-08-02
WASTE: Run Kimi K3's 2.78T Params on 29GB RAM
WASTE is a dependency-free C engine running Kimi K3 (2.78T params) on 29GB RAM by streaming MoE experts from NVMe. Covers real throughput, setup, and limits.
Kimi K3Moonshot AIMoE+4
Business DX2026-08-01
SEO Service Costs: A ¥50K Plan vs. a ¥500K One
A neutral guide to SEO costs by service type, why prices vary so widely, vendor red flags to avoid, and a checklist worth reviewing before signing anything.
Web制作費用中小企業DXSEO
AI2026-08-01
K-EXAONE 2.0 750B-A37B: Self-Hosting a 750B MoE (Apache 2.0)
LG AI Research's K-EXAONE 2.0 750B-A37B needs ~1.5TB weights at BF16, ~750GB at FP8, and 375-420GB at 4-bit — hardware math and Apache 2.0 self-host limits.
K-EXAONELG AI ResearchOpen Weight LLM+4
AI2026-08-01
Mathematica 15: Built-in AI Assistant & LLM Integration
Wolfram Research released Mathematica Version 15 on June 16, 2026, adding an AI Assistant chatbar to every notebook and letting AI call Wolfram's computation.
MathematicaAI
AI2026-08-01
Lean 4 Explained: The Theorem Prover Powering AI Math
Lean 4 is a proof assistant that verifies proofs by compiling them. In 2026, AI-Lean pairs such as AlphaProof Nexus and Gauss delivered striking new results.
LeanAI
Business DX2026-07-31
Files Deleted from a Shared Folder: First Response Guide
What to do first when a file vanishes from a shared folder or cloud drive: what to stop immediately, recovery options and deadlines by storage type (Windows shares, NAS, Microsoft 365, Google Drive), actions that destroy your chances, and who to call.
IT緊急対応中小企業DX保守運用
AI2026-07-31
GPT-5.6 API Price Cut (July 30, 2026): Luna −80%, Terra −20%, and Fast Mode for Sol
On July 30, 2026 OpenAI cut GPT-5.6 API prices: Luna drops 80% to $0.20 in / $1.20 out per 1M tokens, Terra drops 20% to $2 / $12, Sol is unchanged and gains Fast mode (up to 2.5x speed at 2x price). Full old-vs-new tables, monthly cost recalculations, and what to change in your routing design.
GPT-5.6OpenAIAPI料金+1
AI2026-07-30
Qwen Scribe Explained: Fully Local Transcription and System-Wide Dictation on Apple Silicon
Qwen Scribe runs Qwen3-ASR on Apple Silicon via MLX for fully offline transcription and system-wide dictation — 3.4GB unified memory for the 1.7B model, 1.2GB for 0.6B. Covers requirements, setup, push-to-talk usage, SRT export, the three macOS permissions, and how it compares to Whisper-based tools.
ローカルLLM音声入力音声認識+2
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051