<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>News — Nicholas Cabel</title><link>https://news.nicholascabel.com/</link><description>Short stories for students, on money, and on AI.</description><language>en-us</language><atom:link href="https://news.nicholascabel.com/feed.xml" rel="self" type="application/rss+xml"/><item><title>Anthropic ships Claude Fable 5.1 with 75% cheaper cache reads</title><link>https://news.nicholascabel.com/story/2026-09-10-claude-fable-5-1/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-claude-fable-5-1/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>AI</category><description>AI summary of reporting by VentureBeat. Anthropic put Claude Fable 5.1 on general release September 1, leaving base API prices alone but slashing the cost of a cache hit by three quarters, and it reports sizable jumps over Fable 5 on agentic research, coding and business-workflow benchmarks.</description></item><item><title>Cursor lets cloud agents run on machines you control</title><link>https://news.nicholascabel.com/story/2026-09-10-cursor-self-hosted-machines/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-cursor-self-hosted-machines/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>AI</category><description>AI summary of reporting by Cursor. Cursor&#x27;s Sep 2 changelog extends self-hosted machines, which run an agent&#x27;s tool calls on hardware inside your network, with support for outside sandbox hosts and computer use on Linux and Mac workers.</description></item><item><title>DeepSeek-V4.1-Flash lands: MIT weights, 1M context, cheaper Flash rates</title><link>https://news.nicholascabel.com/story/2026-09-10-deepseek-v4-1-flash/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-deepseek-v4-1-flash/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>AI</category><description>AI summary of reporting by The Next Web. DeepSeek released V4.1-Flash, a 552B MoE that activates 8B/16B params, under MIT with 1M context and cut Flash API rates; it has since dropped a plan to route all V4-Pro traffic to it from Sept 14.</description></item><item><title>2027-28 FAFSA is in open beta now, full launch by Oct. 1</title><link>https://news.nicholascabel.com/story/2026-09-10-fafsa-2027-28-beta/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-fafsa-2027-28-beta/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>Students</category><description>AI summary of reporting by Federal Student Aid (U.S. Department of Education). Any student or parent can now request early access to the 2027-28 FAFSA at StudentAid.gov/joinbeta; a beta filing is a real one, and the form reaches everyone no later than Oct. 1, 2026.</description></item><item><title>Google ships Gemini 3.5 Transcribe with batch and live speech-to-text models</title><link>https://news.nicholascabel.com/story/2026-09-10-gemini-3-5-transcribe/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-gemini-3-5-transcribe/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>AI</category><description>AI summary of reporting by Google (The Keyword). Google&#x27;s gemini-3.5-transcribe and gemini-3.5-transcribe-live are in public preview; Google relays an Artificial Analysis WER of 2.6% for the batch model, lists 85+ languages, and its pricing page pegs batch transcription at roughly half a cent per minute.</description></item><item><title>Google ships Gemini 3.8 Flash and a gated Flash Cyber twin</title><link>https://news.nicholascabel.com/story/2026-09-10-gemini-3-8-flash/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-gemini-3-8-flash/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>AI</category><description>AI summary of reporting by Google (The Keyword blog). Google&#x27;s Gemini 3.8 Flash, its third Flash model inside six weeks, keeps 3.7 Flash&#x27;s $0.75/$3.75 intro pricing while a Cyber variant ships only to vetted defenders.</description></item><item><title>GitHub will bill card and PayPal customers up front for Copilot Business and Enterprise seats</title><link>https://news.nicholascabel.com/story/2026-09-10-github-copilot-billing/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-github-copilot-billing/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>AI</category><description>AI summary of reporting by GitHub Changelog. GitHub&#x27;s Aug 28 changelog lays out three Copilot changes: up-front seat billing on Business and Enterprise plans for customers who pay by card or PayPal, one unified chat and cloud-agent experience, and a heavier default effort for Copilot code review.</description></item><item><title>Z.ai&#x27;s GLM-5.3-Flash is the open-weight model behind OpenRouter&#x27;s Ox Alpha</title><link>https://news.nicholascabel.com/story/2026-09-10-glm-5-3-flash/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-glm-5-3-flash/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>AI</category><description>AI summary of reporting by SiliconANGLE. Z.ai released GLM-5.3-Flash, a 320B MoE with 18B active under MIT, confirmed it was OpenRouter&#x27;s mystery Ox Alpha, and priced it at about a tenth of GLM-5.2.</description></item><item><title>Google gives U.S. college students a year of AI Pro free, worth about $240</title><link>https://news.nicholascabel.com/story/2026-09-10-google-ai-pro-free-for-students/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-google-ai-pro-free-for-students/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>Students</category><description>AI summary of reporting by Google (The Keyword blog). Google&#x27;s back-to-school offer hands eligible U.S. college students 12 months of Google AI Pro, a plan it lists at $19.99 a month, as long as they claim it by Dec. 31, 2026; the roughly $240 figure is simply that rate times twelve.</description></item><item><title>OpenAI&#x27;s GPT-6 Astra: 1.05M context, $10/$50, first Critical-tier cyber model</title><link>https://news.nicholascabel.com/story/2026-09-10-gpt-6-astra/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-gpt-6-astra/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>AI</category><description>AI summary of reporting by OpenAI (GPT-6 Astra system card). OpenAI&#x27;s gpt-6-astra brings a 1.05M-token window, 128K output and $10/$50 pricing, and its system card makes it the first OpenAI model rated Critical for cyber.</description></item><item><title>OpenAI ships GPT-Image-2.5 as two API models, Flare and Sunburst</title><link>https://news.nicholascabel.com/story/2026-09-10-gpt-image-2-5/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-gpt-image-2-5/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>AI</category><description>AI summary of reporting by The Decoder. GPT-Image-2.5 reaches OpenAI&#x27;s API as two models, fast Flare and slower, more precise Sunburst, at the same per-token rates GPT Image 2 charged.</description></item><item><title>Grok 4.6 ships with 500K context and $2 input pricing</title><link>https://news.nicholascabel.com/story/2026-09-10-grok-4-6/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-grok-4-6/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>AI</category><description>AI summary of reporting by VentureBeat. Grok 4.6, from SpaceXAI (formerly xAI), is a frontier model aimed at long agent runs and coding, with a 500K window, $2/$6 per million tokens, and a 61 on Artificial Analysis&#x27;s Intelligence Index at launch.</description></item><item><title>Tencent open-sources Hy4 preview: 770B MoE, 1M context, Apache 2.0</title><link>https://news.nicholascabel.com/story/2026-09-10-hy4-preview/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-hy4-preview/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>AI</category><description>AI summary of reporting by TechNode. Tencent&#x27;s Hy team has put out Hy4 preview, a 770B-parameter mixture-of-experts model that activates 49B per token and handles a 1M-token window, with weights on Hugging Face under Apache 2.0 and paid API access through Tencent Cloud and OpenRouter.</description></item><item><title>IRS shuts FIRE 1099 e-file system Nov. 19; filers must move to IRIS</title><link>https://news.nicholascabel.com/story/2026-09-10-irs-fire-to-iris/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-irs-fire-to-iris/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>Financial</category><description>AI summary of reporting by Internal Revenue Service newsroom. FIRE, the IRS system for e-filing 1099s and other information returns, stops taking filings on Nov. 19, 2026; anyone still on it must move to IRIS before tax year 2026 forms are due.</description></item><item><title>MBZUAI&#x27;s IFM ships K2 Horizon, six Apache 2.0 models up to 375B</title><link>https://news.nicholascabel.com/story/2026-09-10-k2-horizon/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-k2-horizon/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>AI</category><description>AI summary of reporting by The National. IFM released K2 Horizon, six Apache 2.0 models from 0.9B to a 375B MoE with 512K context, and has already posted intermediate checkpoints and five training datasets; its code repository is due by end of September, and hosted access runs through Cerebras, AWS and Nebius.</description></item><item><title>Meta opens Muse Glimmer, a 30B Apache 2.0 agent model for 24GB GPUs</title><link>https://news.nicholascabel.com/story/2026-09-10-muse-glimmer/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-muse-glimmer/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>AI</category><description>AI summary of reporting by VentureBeat. Meta&#x27;s Muse Glimmer is a 29.6B dense model distilled from Muse Spark, released as Apache 2.0 weights and sized to run agents on one 24GB GPU.</description></item><item><title>Muse Spark 1.3 keeps its API price, with its top &#x27;max&#x27; mode added after launch</title><link>https://news.nicholascabel.com/story/2026-09-10-muse-spark-1-3/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-muse-spark-1-3/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>AI</category><description>AI summary of reporting by VentureBeat. Muse Spark 1.3 is in Meta&#x27;s API at an unchanged $1.25/$4.25 per million tokens with 1M context; Meta held back its top &#x27;max&#x27; reasoning mode at launch for more safety testing and now says it is available.</description></item><item><title>Nvidia ships Nemotron 3.5 Lightning, a 30B/3B-active open MoE for agents</title><link>https://news.nicholascabel.com/story/2026-09-10-nemotron-3-5-lightning/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-nemotron-3-5-lightning/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>AI</category><description>AI summary of reporting by NVIDIA Blog. Nvidia has released Nemotron 3.5 Lightning, a 30B-parameter MoE that activates 3B parameters per token, under the OpenMDW-1.1 licence, alongside NeMo Switchyard, an open-source router that directs agent requests to different models.</description></item><item><title>Penn State opens AI Studio to every student this fall, with a privacy promise</title><link>https://news.nicholascabel.com/story/2026-09-10-penn-state-ai-studio/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-penn-state-ai-studio/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>Students</category><description>AI summary of reporting by Penn State News. Penn State&#x27;s fall guide says all students now get AI Studio, which puts Claude, ChatGPT and Gemini behind one login, plus the self-paced AI Essentials course, and that your chat logs won&#x27;t be handed to academic-integrity cases.</description></item><item><title>Qwen3.8-27B puts frontier-class coding in a 17GB Apache 2.0 download</title><link>https://news.nicholascabel.com/story/2026-09-10-qwen3-8-27b/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-qwen3-8-27b/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>AI</category><description>AI summary of reporting by The Next Web. Alibaba&#x27;s Qwen team has released Qwen3.8-27B, a dense, Apache 2.0, multimodal model with 262K native context that shrinks to roughly 17GB at 4-bit and carries vendor-claimed coding scores in frontier territory.</description></item><item><title>Sallie Mae: families spent $34,019 on college in 2025-26, up 10%</title><link>https://news.nicholascabel.com/story/2026-09-10-sallie-mae-college-costs-2026/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-sallie-mae-college-costs-2026/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>Financial</category><description>AI summary of reporting by Business Wire (syndicated on Yahoo Finance). Private student lender Sallie Mae&#x27;s How America Pays for College 2026 report puts the average family outlay for the 2025-26 year at $34,019, roughly 10% more than the $30,837 recorded a year earlier, and finds that a slim majority of families paid less than the sticker price.</description></item><item><title>Sept. 30 deadline: enroll in auto pay for a 1-point student loan rate cut</title><link>https://news.nicholascabel.com/story/2026-09-10-student-loan-autopay-deadline/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-student-loan-autopay-deadline/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>Financial</category><description>AI summary of reporting by Business Insider (syndicated on Yahoo Finance). Switching on auto pay by the end of Sept. 30 takes a full percentage point off the interest rate on federal Direct Loans made since July 2012, and the cut lasts through June 30, 2028.</description></item><item><title>Welcome to the AI desk</title><link>https://news.nicholascabel.com/story/2026-09-10-welcome-to-the-ai-desk/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-welcome-to-the-ai-desk/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>AI</category><description>The AI desk is for short AI summaries of model and tool news, and for my own notes from building with Claude and the Model Context Protocol and from working on L.U.M.I., a desktop assistant written in WPF.</description></item><item><title>Welcome to the Financial desk</title><link>https://news.nicholascabel.com/story/2026-09-10-welcome-to-the-financial-desk/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-welcome-to-the-financial-desk/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>Financial</category><description>Money, written by someone still learning it in public. Studio revenue, the THON pledge, and what a student business actually costs to run.</description></item><item><title>Welcome to the Students desk</title><link>https://news.nicholascabel.com/story/2026-09-10-welcome-to-the-students-desk/</link><guid isPermaLink="true">https://news.nicholascabel.com/story/2026-09-10-welcome-to-the-students-desk/</guid><pubDate>Thu, 10 Sep 2026 12:00:00 +0000</pubDate><category>Students</category><description>What this desk covers, how often it writes, and the one page every Penn State student should bookmark first.</description></item></channel></rss>
