produced by Classmethod

# 生成AI articles

GPT-3やChatGPTなどの大規模言語モデルを活用した生成AIについて紹介しています。

I tried the new GPT-6.1 model "Terra" on Amazon Bedrock

2026.10.27

If you're making Supabase MCP read-only, it's also safer to stop the claude.ai connector too

越井琢巳 (Koshii Takumi)

2026.10.26

[Presentation Material] I talked about "Security Considerations for Development Environments in AI-Driven Development" #devio2026

2026.10.25

I presented at DevelopersIO 2026 Tokyo with the title "Hosting AI Agents on DGX Spark and Collaborating via Slack"

嶋田怜央成 / Reona Shimada

2026.10.24

I gave a presentation titled "Key Points of Harness Engineering in AI-Driven Development" #devio2026

2026.10.24

I tried running the 4GB conversational model "Nemotron 3.5 Bolt E4B" on DGX Spark

森茂洋 / Hiroshi Morishige

2026.10.24

AI Cowork Skill for Supporting As-Is and To-Be Creation to Drive Company-Wide AI Transformation

飯沼　深雪

2026.10.21

I gave a presentation at a Claude Code seminar with the title "Designing from 'When Does It End' ── An Introduction to Loop Engineering"

嶋田怜央成 / Reona Shimada

2026.10.20

Reading OpenAI's misalignment reports: How an internal model restricted to "GET only" exceeded its limits and how it was discovered

2026.10.18

Tried Running the Image-Capable 120B MoE "Nemotron 3.5 Super VL 120B-A12B" on DGX Spark

森茂洋 / Hiroshi Morishige

2026.10.18

Kiro supports Kimi K3, I tried comparing models and artifacts with Claude and GPT

2026.10.17

I ran the 2-bit quantized MoE "Nemotron 3.5 Bolt 74B-A9B" on DGX Spark

森茂洋 / Hiroshi Morishige

2026.10.16

I tried the non-generative diffusion model Nemotron Deciffusion 8B on DGX Spark

森茂洋 / Hiroshi Morishige

2026.10.15

[Strands Decider] Ran it on an M5 Mac (16GB) and measured speed and memory for MPS, MLX, and CPU

けーま (川田圭真)

2026.10.06

I compared document value matching using Cloudflare's Clef with Jev

けーま (川田圭真)

2026.10.05

I tried having Cloudflare's Clef classify document types

けーま (川田圭真)

2026.10.04

I quickly tried making an image generation AI model [GAN]

児玉慶一(ケーイチ) | 生成AIコンサルタント@Classmethod

2026.10.02

[Presentation Slides] Looking Back on AI Utilization by One AWS Engineer #devio2026

2026.09.29

Tried running Pi sub-agents per sandbox with NVIDIA OpenShell 0.1.0 series

森茂洋 / Hiroshi Morishige

2026.09.26

I tried using Nemotron 3.5 Lightning as the judge, making it a single-token judgment that doesn't generate like Jev

森茂洋 / Hiroshi Morishige

2026.09.23

I tried out the newly released Amazon Bedrock AgentCore Runtime V2

2026.09.19

I read OpenJev and thought of an alternative solution using non-generative AI

森茂洋 / Hiroshi Morishige

2026.09.18

I tried implementing a library that integrates TypeSafe (Jev) into NeMo Switchyard as a classifier

森永大志 - モリッシー

2026.09.18

注目の記事

I tried forcibly turning Jev into a language model

2026.09.17

注目の記事

I tried replacing model routing with TypeSafe (Jev) to see how much faster and cheaper it would get

森永大志 - モリッシー

2026.09.17

I measured how much the output volume and time required change with Vercel Labs' agent-browser

越井琢巳 (Koshii Takumi)

2026.09.17

I verified a method for browsing files in Windows shared folders and network drives from Amazon Q on Desktop

2026.09.15

I tried running NeMo AutoModel, which can train Hugging Face models as-is, on DGX Spark

森茂洋 / Hiroshi Morishige

2026.09.15

I tried verifying document-level ACL for S3 knowledge bases in Amazon Bedrock Managed Knowledge Base from Amazon Q - Serving confidential documents differently for each user

2026.09.14

I tried validating document-level ACL for S3 knowledge bases in Amazon Q - Serving confidential documents differently per user

2026.09.13

Claude Code v2.1.267 Major Updates - Addition of maxEffortLevel and Comprehensive Fixes Around Prompt Caching

2026.09.10

I thought about AI agent Memory from the perspectives of correction, inspection, and deletion

森茂洋 / Hiroshi Morishige

2026.09.07

[Limited seats remaining] We will be holding "DevelopersIO 2026 SENDAI" in Sendai on Friday, 9/18! #devio2026

2026.09.07

ちょっと話題の記事

I compared Qwen3 8B and GLM Z1 9B Flash across 4 configurations by use case on DGX Spark

森茂洋 / Hiroshi Morishige

2026.09.06

I thought about a configuration for delivering reasoning models to the field on NVIDIA Jetson

森茂洋 / Hiroshi Morishige

2026.09.05

ちょっと話題の記事

I tried running GLM-5.3-Flash 320B on 2 DGX Spark units and measured whether using 2 units is worth it

森茂洋 / Hiroshi Morishige

2026.09.01