Search “Agents”

OpenRouter: Agent usage up 34x compared with February

OpenRouter said that since early February 2026, agentic usage has grown 34-fold; on a weekly basis, agent usage is currently about 8 times that of human usage.

DeepSeek-V4.1-Flash Tops Arena's Open-Source Model Leaderboard

Arena announced that DeepSeek-V4.1-Flash-20261017 (Max) has entered the Arena, ranking first among open-source models in both the Agent Arena and Code Arena: WebDev leaderboards. Arena also mentioned that the 3D world generated by this model, as well as more detailed ranking information, can be viewed in the related post.

English translation is being prepared.

How Octave built a secure multi-tenant AI foundation on Amazon Bedrock

Octave built a three-tier multi-tenant AI foundation on Amazon Bedrock for the quality management platform Reliance, cutting deployment time per use case by 50%. The foundation uses Cohere Embed v4 to embed 2 billion documents, executes Anthropic Claude models through AgentCore Gateway and isolated microVMs, and Guardrails filters content on both the input and output sides. Reliance serves more than 650 enterprise customers across 180 countries.

Simon Willison built Battle Drones, an audio-networked RTS game, with Claude Code

Simon Willison released Battle Drones, a real-time strategy game developed entirely through vibe coding with Claude Code. It went through 12 major versions and 321 commits and runs in mobile and desktop browsers. The game includes 9 building types, 7 ground vehicles, 6 air vehicles and 3 infantry units, and supports fog of war, random map generation, AI versus AI, and an Agent battle mode in which two LLMs fight each other. He also implemented a multiplayer method in which two devices synchronize game state via microphone audio, and because the audio caused headaches, he added online multiplayer based on a WebSockets relay server.

English translation is being prepared.

IRONSCALES Report: AI Agents Are Becoming the New Target of Phishing

The "From Deepfakes to Double Agents" report, published by IRONSCALES after commissioning Censuswide to survey 501 U.S. IT and security practitioners, shows that 77% of respondents said their organization's AI agents or tools had processed malicious, misleading, or unauthorized content in the past year, and 57% reported at least one confirmed incident. 96% of surveyed organizations already use AI tools that can perform actions on their behalf, of which 67% allow them to access ticketing or help desk systems and 49% allow access to email and messaging systems, while 78% of such tools are approved under the leadership of IT departments, with only 6% led by security departments.

English translation is being prepared.

English translation is being prepared.

Salesforce Announces $5.66 Billion Acquisition of Marketing Automation Company Braze

Salesforce announced it will acquire marketing automation company Braze for $5.66 billion in cash, or $47.5 per share, a premium of about 58% over Braze's previous closing price. Braze's revenue grew 26% year over year last quarter to $227.2 million; its platform aggregates customer data through more than 190 prebuilt connectors and offers BrazeAI to automatically generate marketing materials and Agentic Standards to set constraints for AI-generated campaigns. Salesforce plans to use its technology to enhance Marketing Cloud's cross-channel personalization capabilities, and the deal is expected to close in its fiscal year 2028.

English translation is being prepared.

ElevenLabs Releases Two Real-Time Speech-to-Speech Models, Eleven Realtime

ElevenLabs has released two speech-to-speech models, Eleven Realtime, among which v1 Max scores 98.8% on Big Bench Audio with a response time of 0.57 seconds. The smaller v1 Mini targets low-cost natural conversation and can tell whether the other party has finished speaking or only said "mm-hmm," while v1 Max targets more complex tasks such as customer service. Both models are available through the ElevenAgents platform and the OpenAI-compatible Realtime API, billed per minute plus tokens, with v1 Max priced higher than v1 Mini.

English translation is being prepared.

English translation is being prepared.

How AWS Uses X.509 PKI to Secure AI Agent Identity

AWS and Keyfactor jointly released a reference architecture that uses Keyfactor EJBCA's X.509 PKI to establish verifiable, revocable cryptographic identities for AI agents, filling the gap left by MCP's lack of a built-in authentication standard. The solution combines Amazon Bedrock, the AWS Strands Agents SDK, AWS Secrets Manager, and Amazon RDS, and covers three trust boundaries—user authentication, agent-to-agent communication, and MCP tool invocation—through four layers of mechanisms: signed request envelopes, user-signed delegation assertions, RFC 8693 audience-restricted tokens, and real-time OCSP revocation checks.

Meta's Open-Weight Model Muse Spark Open Launches on Fireworks Serverless API

Fireworks launched Meta's open-weight model Muse Spark Open on the day of release. It is an 816B-parameter MoE model that activates 42B parameters per token, with a 1M-token context window, targeting long-horizon coding and agentic tasks. Fireworks says it ranks first among 75 models on Vals AI's Harvey Legal Agent Benchmark leaderboard with 25.4%, nearly 4 times that of Claude Opus 5 (6.7%); on the Rogo Big Finance Bench it surpasses Claude Opus 5.5 at every difficulty level. On Fireworks, the model has zero data retention enabled by default, uses only US-based compute at no extra charge, and supports fine-tuning with serverless LoRA.

Daytona Raises $60M Series B

Daytona has raised a $60M Series B at a $1B valuation, led by FirstMark Capital, with the goal of building the "Agent Cloud." Matt Turck said FirstMark led the Series A earlier this year and has now led the Series B as well, adding that the Daytona team is building one of the defining companies of the AI era.

English translation is being prepared.

English translation is being prepared.

Stanford CME 295 Lecture 4 released: Teaching reinforcement learning with LLMs

The Stanford course CME 295 Transformers & Large Language Models has been updated to Lecture 4, Reinforcement learning with LLMs, and both the video and lecture notes are now public. This lecture goes from RL and Reward Design to PPO, GRPO, RLVR, and DAPO, explaining the notation in each one by one. The course is taught by Afshine and Shervine Amidi, with 9 lectures in total, covering Transformer architecture, attention variants, RoPE, LoRA fine-tuning, RLHF/GRPO training, Flash Attention inference optimization, MCP and AI Agents, and diffusion language models.