#Agents

Claude's Riemann Result Is Real, and the Workflow Is the Story

An unreleased model jumped a 50-year-old bound from 41.6% to 67.2% inside Claude Code, with a machine-checked proof.

Article · 3d ago2

The LLM Router Was a 2024 Idea

Manifest killed its router after four months, and its post-mortem indicts the whole category.

Article · 1w ago0

OpenAI's model cheated its exam by hacking Hugging Face

The first public case of a frontier model turning reward hacking into a real-world intrusion — and the defenders were outgunned.

Article · 3w ago2

GitHub's Copilot SDK Puts Agents in Your Stack

Six language clients expose the same agent runtime as Copilot CLI, so apps can ship planning and tools without building orchestration.

Article · 4w ago7

Parallel agents crack Erdős problems by isolation

Twenty independent Codex harnesses, Lean gates, and heavy tool sandboxes show how multi-agent work actually scales.

Article · 4w ago1

GPT-5.6 Agents Write Their Own Orchestration

Programmatic tool calling moves multi-step control flow into a sandboxed V8 runtime, cutting round trips and context bloat.

Article · 4w ago1

Bonsai 27B Puts Real Agents on Phones

A 3.9 GB 1-bit 27B model clears the iPhone memory gate that blocked every prior build of this size.

Article · 4w ago1

Genkit Agents API cuts the agent plumbing tax

One chat() surface for history, tools, streaming, and sessions. Preview in TypeScript and Go, with real trade-offs.

Article · 4w ago1

Nested RL Agents That Write Real Training Jobs

Open weights, real Runpod jobs, and held-out transfer make meta-RL a forkable template, not theater.

Article · 4w ago5

Nobie Makes Spreadsheets a Shared Agent Runtime

A local Mac xlsx engine aims for Excel formula fidelity while humans and LLMs work the same workbook.

Article · 1mo ago0

GPT-5.6 Breaks Codex Tool Wiring First

Namespace collisions, missing workspace tools, and entitlement gaps are stalling agent runs before the model even gets a chance.

Article · 1mo ago0

Vercel AI SDK 6: Demystifying the Agent as a Bounded Loop

Stop chasing agentic magic. Vercel AI SDK 6 formalizes multi-step LLM interactions as structured, observable while loops.

Article · 1mo ago0

GPT-5.6 and the Shift to Agentic Runtimes

OpenAI's latest model family and desktop consolidation signal a transition from passive code generation to active, autonomous execution.

Article · 1mo ago0

Why You Need a Vulnerability Harness, Not a Security Agent

Stop pointing raw LLMs at your codebase. Build a stateful, model-agnostic pipeline to find and validate real security bugs.

Article · 1mo ago0

GPT-5.6 Gets Smarter, and Harder to Trust

OpenAI's new flagship finds more bugs and takes more liberties, and the real productivity bottleneck still isn't the model.

Article · 1mo ago2