We read it. We curate it. Model releases, tools, papers, and hands-on tutorials — filtered so you don't have to scroll through the noise.

code(3) learning(14) ml ops(2) models(1) news(5) papers(5) security(3)

Building a Memory-Driven Agent with NVIDIA NemoClaw

2026-09-13

A tutorial for building a memory-driven "self model" agent that scores 90.9%, compared with 82.8% for an agentic-RAG baseline, while changed-fact tracking improves from 60% to 100%.

learning

Fine-tuning a 350M Model for Better Structured Outputs in 100 GRPO Steps

2026-09-13

A hands-on guide fine-tunes a 350M model on a free 16GB GPU, improving JSON format compliance from 18.0% to 31.9% in 100 GRPO steps.

learning

Detecting and countering misuse of AI: September 2026

2026-09-13

Anthropic's Threat Intelligence team identified and disrupted multiple operations that used Claude for malicious activity between December 2025 and August 2026. The report presents case studies and tracks how misuse has evolved, noting that none involved Claude Fable or Mythos-class models.

learningsecurity

WearableQA Benchmarks Health Reasoning on Real-World Wearable Data

2026-09-13

WearableQA provides 4,084 ten-option questions built from wearable time series, blood biomarkers, and demographics from 200 real users. The benchmark tests models on longitudinal data analysis and physiological reasoning, with raw structured data and customizable renderings included.

code

Terminal-Universe: Turning Agent Trajectories into Scalable Terminal Environments

2026-09-13

A new framework reconstructs reusable coding environments from existing terminal-agent trajectories instead of creating them from zero. It restores workspaces, fills missing files, and generates original, cross-project, and multi-round tasks. From public trajectories, it builds 37.3k environments and improves coding-agent performance on two benchmarks.

papers

AI Has Solved One of Math's $1 Million Millennium Prize Problems

2026-09-12

OpenAI deployed 10,000 collaborating agents for 88 hours to produce a Lean-verified proof of a finite-time singularity in the 3D Navier-Stokes equations, shortly after an independent competing announcement.

news

AlphaGenome Atlas: A predictive map of every possible DNA letter change in the human genome

2026-09-12

AlphaGenome Atlas scores all 9 billion possible single-letter DNA variants in the human genome and is available free for noncommercial research.

news

Q2D-Web: Evaluating First-Stage Retrievers at Scale

2026-09-12

A multilingual benchmark and leaderboard evaluates retrieval models across 190 million documents and 69,721 queries. Independent relevance judgments and cost-efficient subsampling improve evaluation reliability while preserving model rankings.

learning

Data bottlenecks won't prevent an intelligence explosion

2026-09-12

More sample-efficient learning algorithms could reduce AI's dependence on vast datasets. Even if data quality proves difficult to improve beyond human levels, data constraints may not prevent AI from automating most economic work.

learning

Linguistic drift at the frontier

2026-09-12

An analysis of 685,000 GitHub pull requests traces Claude-popularized vocabulary and presents VocabGuard, a drift-monitoring system with 0.869 AUC.

learning

Look, But Don't Touch (Read-Only Tools for AI Agents)

2026-09-12

MCP Toolbox for Databases uses protocol-level controls to enforce read-only access, avoiding prompt guardrails, regex parsers, and session flags that can fail under injection.

learningsecurity

Long-context latency scales quadratically for GPT-5.6 but nearly linearly for Claude 5

2026-09-12

Analysis of time-to-first-token latency up to million-token contexts finds quadratic scaling for GPT-5.6 and nearly linear scaling for Claude 5.

learning

How llm-d makes the most of the hardware you already have

2026-09-12

A benchmark shows llm-d serving a 753B-parameter mixture-of-experts model across 544 H100 GPUs for 3,000 concurrent coding agents, with 85.2% of input tokens served from cache.

ml ops

OpenCodeReview: Precise AI-Powered Code Reviews

2026-09-12

OpenCodeReview is a CLI tool combining deterministic pipelines with an LLM agent to review Git diffs and entire files. It delivers line-level comments, supports multilingual rules for issues such as NPE, thread safety, XSS, and SQL injection, and works with OpenAI- and Anthropic-compatible providers.

code

How Fast Do Agents Rot? An Empirical Study of Long-Horizon Degradation in LLM Agents for Production Decision-Making

2026-09-12

A study of nine LLMs finds that agent success drops geometrically as the number of dependent steps grows. Even strong models can fall from near-perfect to near-zero success within 16 steps, mainly because of step-level errors rather than context length.

papers

Fractal basins trap latent reasoning

2026-09-12

AI reasoning models slow down on harder problems because their internal process can enter transient chaotic states. They become trapped near nearly correct solutions, creating fractal patterns in how they search across tasks such as Sudoku, mazes, visual puzzles, and mathematical logic.

papers

Introducing Mercury 2.5

2026-09-11

Mercury 2.5 delivers 1,107 tokens per second on widely available Nvidia GPUs, a 260K-token context window, and performance comparable to cost-optimized frontier models. Launch pricing starts at $0.04 per million input tokens and $0.15 per million output tokens, with an 80% introductory discount.

news

Google's AlphaGenome Maps 9 Billion Genetic Variants

2026-09-11

AlphaGenome Atlas is a 1-petabyte database predicting the regulatory effects of all 9 billion possible single-nucleotide variants in the human genome.

news

Model Card for North Small Translate

2026-09-11

North Small Translate is an open-weights research model with 25 billion active parameters and 218 billion total parameters, specialized for high-quality machine translation across 50 languages.

newsmodels

Pretraining progress is mostly coming from data

2026-09-11

From 2019 to 2025, data improvements delivered 3.24 times more compute-efficiency gains than model improvements. The gains were largely independent, while model research mainly removed scaling constraints; data quality matters most for smaller models and may matter less at larger scales.

learning

Hyper-𝜏-bench: Evaluating agents that build agents

2026-09-11

A benchmark puts developer agents in simulated business environments to recover requirements, design architectures, build tools, and deploy a customer-service agent within a model-cost budget. Claude Opus 5 scores 23.9% alone versus 82.2% when paired with an experienced engineer.

learning

Inside the megakernel serving engine for North Mini Code

2026-09-11

A fully fledged serving system combines continuous batching, paged attention, ragged sequence lengths, tool calling, and an OpenAI-compatible endpoint. The megakernel reaches 292 tokens per second at batch size 1, outperforming vLLM by 1.58x across batch sizes and up to 256K-token contexts without measurable accuracy loss.

learning

GPT-6 Astra, Looped Transformers, and Hidden Reasoning

2026-09-11

GPT-6 Astra makes a major leap in computer use. A variant of the looped transformer may improve performance at a fixed compute budget, while shorter reasoning traces may reflect fewer mistakes and more internal computation.

learning

Inside the LLM Call: GenAI Observability with OpenTelemetry

2026-09-11

OpenTelemetry's GenAI semantic conventions standardize traces for agent invocations, LLM calls, and tool executions, including token-usage and duration metrics.

learningml ops

Anthropic Finds Claude Misalignment in Cybersecurity Tests

2026-09-11

Four misconfigured cybersecurity evaluations gave Claude access to real systems, according to Anthropic's assessment.

learningsecurity

OKF Agent Memory: Git-Native Persistent Memory for AI Agents

2026-09-11

A pure Go, zero-dependency memory layer that stores AI agent knowledge as version-controlled Markdown and YAML in a repository. It provides OKF v0.2 support, fast local BM25 search, progressive disclosure, validation, CLI tools, and an embedded MCP server without external databases.

code

Long-Horizon Language Model Reinforcement Learning via Progressive Point Matching

2026-09-11

A new reinforcement learning method gives language models dense rewards for matching progress across task segments, making learning more efficient on long and complex tasks. Tests on difficult math problems show that segment-level rewards improve success rates when models use larger test-time token budgets.

papers

The Embedder's Dilemma: LLMs Are Better, but at What Cost?

2026-09-11

A benchmark of 10 LLMs and 26 embedding models across 37 tasks finds similar overall quality. LLMs perform better on reasoning-heavy retrieval, while embedding models are stronger for classification and much cheaper and faster.

papers
For more updates, visit our dedicated website: https://news.boringai.tech