CompanyAnthropic8 recent entries27 Sept 20262026 in LLMs (so far)On Friday I gave the closing keynote at the WeAreDevelopers World Congress North America in San Jose. I tied together the key trends from the past year into a chronological exploration of everything t→28 Sept 2026Qwen3-VL 8B on a laptop vs Opus 5.5 / Sonnet 5 / GPT-5.6 on 137 messy documents: beat GPT-5.6 on tax forms, lost badly on Indian date formats[R]I benchmarked Qwen3-VL 8B Instruct (Q4_K_M, Ollama, M5 24GB, ~30s/doc) against Claude Opus 5.5, Sonnet 5 and GPT-5.6 Terra on: - receipts: CORD (Indonesia) and SROIE (Malaysia), 30 each - 20 scanned 1→
CompanyOpenAI8 recent entries28 Sept 2026Sources on how SoftBank's $11B junk bond sale landed, despite investor questions about OpenAI's listing timeline, data center plans, and SB Energy's delayed IPO (Bloomberg)Bloomberg: Sources on how SoftBank's 11B junk bond sale landed, despite investor questions about OpenAI's listing timeline, data center plans, and SB Energy's delayed IPO — SoftBank Group Corp.'s bill→28 Sept 2026AI is supercharging hacking, and your local hospitals and banks aren’t readyIn March, Janice Malone began getting calls about suspicious activity from her nonprofit organization, Vivian's Door. Vivian's Door, headquartered in Alabama, typically provided training, resources, a→29 Sept 2026OpenAI announces new features for Codex, including reusable cloud development environments, a refreshed Codex CLI, a new code review experience, and more (Sarah Perez/TechCrunch)Sarah Perez / TechCrunch: OpenAI announces new features for Codex, including reusable cloud development environments, a refreshed Codex CLI, a new code review experience, and more — OpenAI introduced →3 Oct 2026Why Doesn’t ChatGPT Warn You Before Wasting 30 Minutes of Compute?I've been using GPT-6 Astra to create 70+ page documents from 100,000+ characters of source material, and it's jump in processing is great. Here's the weird part, when I give it a massive task it work→3 Oct 2026OpenAI's DevDay 2026 announcements to turn ChatGPT into a place to discover, launch, and use software could potentially disrupt the traditional app store model (Sarah Perez/TechCrunch)Sarah Perez / TechCrunch: OpenAI's DevDay 2026 announcements to turn ChatGPT into a place to discover, launch, and use software could potentially disrupt the traditional app store model — The focus of→4 Oct 2026Wiki Lint Report — 2026-10-04Automated lint: 19 errors, 64915 warnings, 3 info→5 Oct 2026Evaluation format, not model capability, drives measured triage failure in the assessment of consumer health AIarXiv:2603.11413v4 Announce Type: replace-cross Abstract: A recent Nature Medicine study reported that ChatGPT Health under-triages 51.6% of emergencies and concluded that consumer-facing AI triage po→8 Oct 2026Can you trust Meta’s Muse or OpenAI’s Dots to run your life?My Decoder guest today is Hayden Field, The Verge’s senior AI reporter, and we’re discussing the new wave of consumer-friendly AI agents. If you’ve been paying attention to this space, you know AI ent
CompanyGoogle8 recent entries23 Sept 2026stuntd: a local Jev-compatible server on Laya that learns from your own traffic (no API key needed)What it is. stuntd is a local proxy for typed LLM decisions: choice, yes/no, score. Your app keeps calling its provider through it. stuntd records the decisions, trains a small head on top of Laya (th→27 Sept 2026Wiki Lint Report — 2026-09-27Automated lint: 17 errors, 59669 warnings, 3 info→27 Sept 20262026 in LLMs (so far)On Friday I gave the closing keynote at the WeAreDevelopers World Congress North America in San Jose. I tied together the key trends from the past year into a chronological exploration of everything t→3 Oct 2026Why Doesn’t ChatGPT Warn You Before Wasting 30 Minutes of Compute?I've been using GPT-6 Astra to create 70+ page documents from 100,000+ characters of source material, and it's jump in processing is great. Here's the weird part, when I give it a massive task it work→4 Oct 2026Wiki Lint Report — 2026-10-04Automated lint: 19 errors, 64915 warnings, 3 info→5 Oct 2026Introducing Google Cloud Modernize, transforming for (and with) AIToday, we’re announcing Google Cloud Modernize, an end-to-end transformation portfolio to help enterprises collapse multi-year roadmaps with the power of AI. Google Cloud Modernize brings together Goo→8 Oct 2026Welcome to Gemini at Work 2026: Introducing the Gemini agentEditor’s note: This article is adapted from Thomas Kurian’s keynote address at Gemini at Work 2026. It includes a number of announcements: Gemini, a single, universal agent for work that answers your →8 Oct 2026Another great paper from Salesforce AI Research. The finding is that a 31B open Gemma-4 web agent scores 74.6% on the 9-app WebArena Infinit…Another great paper from Salesforce AI Research. The finding is that a 31B open Gemma-4 web agent scores 74.6% on the 9-app WebArena Infinity set, above Gemini 3 Flash with browser use at 70.1%. They
CompanyMeta8 recent entries30 Sept 2026Learning Meta-Skills for Agent Harness Design in Test-Time AI4AIarXiv:2609.38143v1 Announce Type: new Abstract: Agent performance depends on both reasoning ability and the environment in which it acts. We study test-time AI-for-AI, asking how a Builder can learn t→1 Oct 2026Outsourcing AI regulation to a voluntary 'pledge' conceived by Mark Zuckerberg and endorsed by tech CEOs over a private White House lunch is…Outsourcing AI regulation to a voluntary 'pledge' conceived by Mark Zuckerberg and endorsed by tech CEOs over a private White House lunch is a recipe for disaster. 'Self-regulation' by tech giants is →3 Oct 2026OpenAI's DevDay 2026 announcements to turn ChatGPT into a place to discover, launch, and use software could potentially disrupt the traditional app store model (Sarah Perez/TechCrunch)Sarah Perez / TechCrunch: OpenAI's DevDay 2026 announcements to turn ChatGPT into a place to discover, launch, and use software could potentially disrupt the traditional app store model — The focus of→4 Oct 2026Synthesis: TechmemeAuto-generated synthesis of 3157 entries about techmeme→4 Oct 2026Wiki Lint Report — 2026-10-04Automated lint: 19 errors, 64915 warnings, 3 info→5 Oct 2026From Benchmarks to Production: A Text-to-SQL System for Complex Financial DataarXiv:2610.03524v1 Announce Type: new Abstract: General-purpose Text-to-SQL systems achieve strong performance on academic benchmarks like Spider and BIRD, where schemas are relatively shallow and col→8 Oct 2026Can you trust Meta’s Muse or OpenAI’s Dots to run your life?My Decoder guest today is Hayden Field, The Verge’s senior AI reporter, and we’re discussing the new wave of consumer-friendly AI agents. If you’ve been paying attention to this space, you know AI ent→9 Oct 2026NOMOS: Compiling Written Policies into Statically Verified Tool-Call Gates for LLM AgentsarXiv:2610.11030v1 Announce Type: cross Abstract: Tool-using LLM agents violate the policies they are deployed to enforce, often silently. Prior defenses hand-write rules, query an LLM verifier per ac
CompanyMistral8 recent entries15 Jul 2026Wiki Lint Report — 2026-07-15Automated lint: 26 errors, 6728 warnings, 3 info→19 Jul 2026Wiki Lint Report — 2026-07-19Automated lint: 20 errors, 8743 warnings, 3 info→3 Aug 2026I gave five different local LLMs a town. They invented Facebook and a duck-based credit bureau. (MIT, self-hosted, you don't play it — you watch it)Each villager in Pepperton is a different model — a mistral, a qwen3, a qwen2.5, a phi4-mini, a llama3.2 — because model families have genuinely different temperaments, and the friction between them i→18 Aug 2026Fiber Fingerprints of Hidden Learning-State DynamicsarXiv:2608.15976v1 Announce Type: new Abstract: A learning system can occupy execution states that are indistinguishable under every declared present-behavior readout yet respond differently to future→8 Sept 2026Mistral raised a €3B Series D led by Samsung at a €21B valuation, up from €11.7B a year ago, as it expands into data centers beyond developing AI models (Adam Satariano/New York Times)Adam Satariano / New York Times: Mistral raised a €3B Series D led by Samsung at a €21B valuation, up from €11.7B a year ago, as it expands into data centers beyond developing AI models — Mistral is t→15 Sept 2026CrofAI 'cheapest inference provider in the world' gets exposed as an OpenRouter wrapper, routing requests to smaller, cheaper models at up to 20x markup. CrofAI responds to Wire Fraud allegations by denying everything, then backtracking, then 3 hours later wiping their entire online presenceDisclaimer: no AI was used whatsoever to write this post Cautionary tale about chasing cheap tokens. exposé: https://kendell.dev/blog/crofaifalse/ reaction by nahcrof, announcing the shutdown of the s→18 Sept 2026Mistral and other European AI startups accuse US rivals of using safety concerns to entrench their dominance, rejecting Anthropic's calls to pace the frontier (Reuters)Reuters: Mistral and other European AI startups accuse US rivals of using safety concerns to entrench their dominance, rejecting Anthropic's calls to pace the frontier — Europe's leading AI company Mi→30 Sept 2026Follow-up: my native Rust + Vulkan Transformer training backend — 14 days later, now 14 parity-verified architectures and full PEFTFollow-up to my post from about two weeks ago. A lot has changed since then, so I wanted to post an update on where the backend is now. Where the green architectures stand When I posted last time, 7 a
CompanyxAI8 recent entries9 Sept 2026Wiki Lint Report — 2026-09-09Automated lint: 12 errors, 53990 warnings, 3 info→13 Sept 2026Wiki Lint Report — 2026-09-13Automated lint: 12 errors, 54839 warnings, 3 info→18 Sept 2026Grok Bot summary of SpaceX CFO Bret Johnsen’s new interview with Tony Kim at SpaceX Mission Control. Bret’s path - Came from semiconductors …Grok Bot summary of SpaceX CFO Bret Johnsen’s new interview with Tony Kim at SpaceX Mission Control. Bret’s path - Came from semiconductors (Qualcomm, Broadcom; was CFO of MindSpeed) - Barely knew Spa→20 Sept 2026Wiki Lint Report — 2026-09-20Automated lint: 14 errors, 54840 warnings, 3 info→27 Sept 2026Wiki Lint Report — 2026-09-27Automated lint: 17 errors, 59669 warnings, 3 info→29 Sept 2026🚨 JUST NOW: Elon Musk, shoulder to shoulder with President Trump, explains why AI is going to lift up ALL of humanity -- not force them int…🚨 JUST NOW: Elon Musk, shoulder to shoulder with President Trump, explains why AI is going to lift up ALL of humanity -- not force them into poverty @elonmusk: 'I think by FAR the most likely outcome →4 Oct 2026Wiki Lint Report — 2026-10-04Automated lint: 19 errors, 64915 warnings, 3 info→8 Oct 2026Last week some of South Korea's biggest banks were hit by a cyberattack. We now know the entire hack may have been done by a single person. He used a combined stack of an open-source AI penetration tool named ARTEX, DeepSeek v4.1-Flash, GLM-5.3, Grok 4.6, and Claude Code (CrowdStrike)CrowdStrike report: https://www.crowdstrike.com/en-us/blog/unknown-threat-actor-uses-artex-to-target-south-korean-finance/ From: Andrew Curran on 𝕏: https://x.com/AndrewCurran_/status/2108108695876092
CompanyDeepSeek8 recent entries3 Sept 2026FUSE: An Evaluating Framework for Dangerous Capabilities of LLMsarXiv:2609.02168v1 Announce Type: new Abstract: Fragmented safety evaluation undermines the governance of dangerous AI capabilities. We present a modular framework that evaluates each model through th→7 Sept 2026New Benchmark: The Struggle BenchHow it works. The model being tested is given a server capable of running it's weights and full context. That server is placed in a median priced apartment. The AI is given a bank account with for ren→10 Sept 2026This is a brilliant paper. It's of the cleanest long-context agent designs I have seen in the past couple of months. Sequential memory agent…This is a brilliant paper. It's of the cleanest long-context agent designs I have seen in the past couple of months. Sequential memory agents read chunks one after another while maintaining a compact →15 Sept 2026CrofAI 'cheapest inference provider in the world' gets exposed as an OpenRouter wrapper, routing requests to smaller, cheaper models at up to 20x markup. CrofAI responds to Wire Fraud allegations by denying everything, then backtracking, then 3 hours later wiping their entire online presenceDisclaimer: no AI was used whatsoever to write this post Cautionary tale about chasing cheap tokens. exposé: https://kendell.dev/blog/crofaifalse/ reaction by nahcrof, announcing the shutdown of the s→30 Sept 2026Follow-up: my native Rust + Vulkan Transformer training backend — 14 days later, now 14 parity-verified architectures and full PEFTFollow-up to my post from about two weeks ago. A lot has changed since then, so I wanted to post an update on where the backend is now. Where the green architectures stand When I posted last time, 7 a→2 Oct 2026Two 96 GB Ascend cards crun Qwen3.8-flash-next hardware notes, vLLM work, benchmarks, and what is nextI have been building a somewhat unusual local inference machine around two Huawei Atlas 300I Duo cards. They are relatively inexpensive, passive, dual-accelerator PCIe cards with 96 GB of device memor→7 Oct 2026ServeLearnBench: How Well Can Agents Self-Improve from Serving Experience?arXiv:2610.07792v1 Announce Type: cross Abstract: Large language model agents are increasingly deployed to perform complex tasks in real-world environments. However, the knowledge required for correct→8 Oct 2026Last week some of South Korea's biggest banks were hit by a cyberattack. We now know the entire hack may have been done by a single person. He used a combined stack of an open-source AI penetration tool named ARTEX, DeepSeek v4.1-Flash, GLM-5.3, Grok 4.6, and Claude Code (CrowdStrike)CrowdStrike report: https://www.crowdstrike.com/en-us/blog/unknown-threat-actor-uses-artex-to-target-south-korean-finance/ From: Andrew Curran on 𝕏: https://x.com/AndrewCurran_/status/2108108695876092
CompanyNVIDIA8 recent entries13 Sept 2026Law enforcers already have authority to charge companies and their CEOs for creating and releasing dangerous, unvetted, or defective product…Law enforcers already have authority to charge companies and their CEOs for creating and releasing dangerous, unvetted, or defective products. We shouldn’t let discussions about new legal regimes dist→14 Sept 2026Nvidia announces the RTX Pro 5500 Blackwell Workstation Edition, offering comparable specs to the RTX 5090 but with 84GB of GDDR7 memory, vs. RTX 5090's 32GB (Zhiye Liu/Tom's Hardware)Zhiye Liu / Tom's Hardware: Nvidia announces the RTX Pro 5500 Blackwell Workstation Edition, offering comparable specs to the RTX 5090 but with 84GB of GDDR7 memory, vs. RTX 5090's 32GB — The GeForce →18 Sept 2026Grok Bot summary of SpaceX CFO Bret Johnsen’s new interview with Tony Kim at SpaceX Mission Control. Bret’s path - Came from semiconductors …Grok Bot summary of SpaceX CFO Bret Johnsen’s new interview with Tony Kim at SpaceX Mission Control. Bret’s path - Came from semiconductors (Qualcomm, Broadcom; was CFO of MindSpeed) - Barely knew Spa→30 Sept 2026GPU cloud provider GMI Cloud raised 668M, including 223M in equity led by ARCHIV with participation from Nvidia and $445M in credit led by Taiwanese bank CTBC (Phoebe Liu/The Information)Phoebe Liu / The Information: GPU cloud provider GMI Cloud raised 668M, including 223M in equity led by ARCHIV with participation from Nvidia and 445M in credit led by Taiwanese bank CTBC — GMI Cloud,→30 Sept 2026Follow-up: my native Rust + Vulkan Transformer training backend — 14 days later, now 14 parity-verified architectures and full PEFTFollow-up to my post from about two weeks ago. A lot has changed since then, so I wanted to post an update on where the backend is now. Where the green architectures stand When I posted last time, 7 a→2 Oct 2026Two 96 GB Ascend cards crun Qwen3.8-flash-next hardware notes, vLLM work, benchmarks, and what is nextI have been building a somewhat unusual local inference machine around two Huawei Atlas 300I Duo cards. They are relatively inexpensive, passive, dual-accelerator PCIe cards with 96 GB of device memor→6 Oct 2026Sources: SpaceX is seeking to raise 40B, including ~10B in bank loans and ~$30B in investment-grade debt, to purchase Nvidia chips, in a deal led by Apollo (Financial Times)Financial Times: Sources: SpaceX is seeking to raise 40B, including ~10B in bank loans and ~$30B in investment-grade debt, to purchase Nvidia chips, in a deal led by Apollo — Blockbuster debt deal is →6 Oct 2026Efficient Analog-Initialized Latent Transport for Spatially Coherent Probabilistic Downscaling from Global to Kilometer ScalesarXiv:2610.03736v1 Announce Type: cross Abstract: Probabilistic downscaling must represent kilometer-scale structure left unresolved by a deterministic regional prediction. We introduce an analog-init