A Nature Computational Science study finds LLMs fail to use coordinate information in materials data regardless of model size, while a Scientific Reports study shows retrieval cutting a minaret risk score from 0.816 to 0.491. What both mean for teams applying language models to science and engineering.
An October 2026 study finds a detector frozen at GPT-4 catches at most 0.6% of GPT-5-era rewrites, while arXiv capped submissions and mathematicians launched Hexagon for AI-generated proofs. What journals, labs and editors should change.
Ai2's Bolmo paper in Nature shows how to retrofit existing language models to read raw bytes instead of tokens. What byteification is, how Bolmo, Bwen and Blama perform, what the two-stage process costs, and which tasks benefit.
Open-weight models now handle most tokens on gateways like Vercel's and OpenRouter, yet closed models still take about 86% of estimated spend. What the September 2026 data shows, why Chinese open models lead, and how a 9B longevity model topped frontier systems on a domain benchmark.
A Nature paper describes Paper2Agent, which converts a research paper and its code into an MCP server that a chat agent can use. How it works, the AlphaGenome case built in about 45 minutes for US$14, accuracy results, limits, and how researchers can use it.
OpenAI's GPT-6 Astra is its first broadly deployed model rated Critical for cybersecurity, and it uses recurrent depth that makes reasoning harder to monitor. A guide to the safety concerns, prompt-injection results, the ARC-AGI-3 score gap, and what Ai2's BenchMIRT says about reading benchmarks.
Google's Gemini 3.8 Flash keeps 3.7 Flash's $0.75/$3.75 introductory price but performs more reasoning steps and tool calls, so tasks can cost more. Benchmarks, the Flash Cyber variant and Fairwind Program, cost per task, and practical guidance on effort levels and model choice.
Google released three new Gemini models this week with meaningful price cuts and benchmark gains. Here is what changed, what the new Flash-Lite tier means for developers, and why the flagship Pro model remains conspicuously absent.
Moonshot's Kimi K3 spooked US markets last week when it beat GPT-5.6 and Claude Fable 5 at coding benchmarks. Now American labs like Poolside, Thinking Machines, and Reflection AI are shipping models that aim to close the gap. The open-source LLM race is no longer theoretical — it's happening, and the results are getting hard to ignore.
Emergent just became the latest AI coding startup to hit a $1B+ valuation, raising $130M barely a year after launch. But the real story isn't the money — it's who's using these tools and why embedded AI might matter more than standalone agents.
A new Nature paper argues our entire understanding of LLMs is built on flawed human metaphors — and proposes a radical alternative called machine experientialism that reframes how we think about AI cognition.
Meta FAIR and academic partners' Brain2Qwerty v2 decodes sentences people type from non-invasive MEG recordings, with 61% average and 78% best-participant word accuracy. How it works, how it compares with v1 and implants, why MEG matters, its limits and the neural privacy questions it raises.
A June 2026 Nature Medicine study found three frontier LLMs outscored two specialized clinical AI tools on MedQA, HealthBench and real clinician queries. The results, the vendor pushback including a retraction demand, and what it means for enterprises choosing general models versus specialist AI.
How chain-of-thought reasoning evolved from a prompting trick (Wei et al., Kojima et al.) through self-consistency, Tree of Thoughts and ReAct to RL-trained reasoning models like o1 and DeepSeek-R1, where it helps, its limits, whether visible reasoning is faithful, and how to prompt reasoning models.
A practical guide to retrieval-augmented generation: how RAG grounds language models in your documents, chunking, hybrid search and re-ranking, agentic and graph-based variants, how to evaluate faithfulness, security and access control, and how RAG compares with long context windows and fine-tuning.
A practical comparison of ChatGPT, Claude and Gemini: published benchmarks, context windows, standout features, ecosystem integration and plan structure, with guidance by role and a simple method for testing them on your own work. Updated for the autumn 2026 model releases.
Anthropic's Claude Opus 4.6 delivers a 1M token context window that actually works, revolutionary agent teams, and benchmark performances that redefine agentic AI capabilities.
From Claude Opus 4.6's million-token context to DeepSeek's memory architecture revolution, explore the technical advances reshaping AI in early 2026.
In-depth analysis of GPT-5, Claude Opus 4.6, and Gemini 3 Pro comparing performance across mathematical reasoning, coding, agentic tasks, and multimodal understanding.
What DeepSeek R1 is and why it mattered: the R1-Zero reinforcement learning experiment, the multi-stage training behind R1, published benchmarks against OpenAI o1, the distilled 1.5B to 70B models, MIT licensing, API pricing at launch and how to run it.
A look back at the pre-launch predictions for GPT-5 and how they compare with what OpenAI actually released on August 7, 2025: a unified system with a router, a 400K-token API context window, and new developer controls.