<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>The Tech Archive</title>
    <link>https://aitecharchive.com</link>
    <description>Deep-dive articles on Azure, AI/LLMs, and C# for software architects and developers.</description>
    <language>en-US</language>
    <lastBuildDate>Thu, 19 Nov 2026 07:47:45 GMT</lastBuildDate>
    <atom:link href="https://aitecharchive.com/rss.xml" rel="self" type="application/rss+xml"/>
    
    <item>
      <title>MAI Code 2 Launch: Price vs Benchmark Reality for Enterprise Coding</title>
      <link>https://aitecharchive.com/articles/mai-code-2-launch-copilot-price-gpt-6-1-sol-benchmarks</link>
      <guid isPermaLink="true">https://aitecharchive.com/articles/mai-code-2-launch-copilot-price-gpt-6-1-sol-benchmarks</guid>
      <pubDate>Thu, 19 Nov 2026 07:29:17 GMT</pubDate>
      <author>admin@techarchive.dev (Sham)</author>
      <category>AI Coding Tools</category>
      <description>Microsoft&apos;s MAI Code 2 hits 98% of GPT-6.1 Sol&apos;s score at 40% of the cost, but the promo price doubles in May 2027. INR rates and GST included.</description>
    </item>
    <item>
      <title>Gemini 4 Neon: $3/$15 Beats Opus 5.5 on Cost per Task</title>
      <link>https://aitecharchive.com/articles/gemini-4-neon-launch-price-benchmarks</link>
      <guid isPermaLink="true">https://aitecharchive.com/articles/gemini-4-neon-launch-price-benchmarks</guid>
      <pubDate>Thu, 19 Nov 2026 05:45:19 GMT</pubDate>
      <author>admin@techarchive.dev (Sham)</author>
      <category>AI Models &amp; Benchmarks</category>
      <description>Google launched Gemini 4 Neon on Nov 18, 2026 at $3/$15 per million tokens with a 1M context. AA Index 61 ranks it #3, and it tops the coding-agent index.</description>
    </item>
    <item>
      <title>Energy-Based Reasoners: 99.9 Percent Sudoku on 345x Less Compute</title>
      <link>https://aitecharchive.com/articles/energy-based-reasoners-999-sudoku-345x-less-compute</link>
      <guid isPermaLink="true">https://aitecharchive.com/articles/energy-based-reasoners-999-sudoku-345x-less-compute</guid>
      <pubDate>Thu, 19 Nov 2026 04:05:35 GMT</pubDate>
      <author>admin@techarchive.dev (Sham)</author>
      <category>AI Research</category>
      <description>A 1.8M-parameter energy-function reasoner from Duke, UIUC and Harvard hits 99.9% on Sudoku-Extreme with 345x less inference compute. MIT code, trains in hours.</description>
    </item>
    <item>
      <title>KARTE Data Breach: PLAID Key Theft Leaks 44 Million Records</title>
      <link>https://aitecharchive.com/articles/karte-data-breach-plaid-keys-leak-japan-44-million</link>
      <guid isPermaLink="true">https://aitecharchive.com/articles/karte-data-breach-plaid-keys-leak-japan-44-million</guid>
      <pubDate>Wed, 18 Nov 2026 20:51:05 GMT</pubDate>
      <author>admin@techarchive.dev (Sham)</author>
      <category>Cybersecurity</category>
      <description>Japan&apos;s KARTE breach: vendor PLAID lost 79,249 logins and integration keys; JINS, DHC and MUFG disclose 44 million records. Timeline, impact and what to revoke.</description>
    </item>
    <item>
      <title>AI Chip Design Goes Production: OpenAI Rewrites AMD RTL</title>
      <link>https://aitecharchive.com/articles/ai-chip-design-openai-amd-rtl-rewrite-synopsys</link>
      <guid isPermaLink="true">https://aitecharchive.com/articles/ai-chip-design-openai-amd-rtl-rewrite-synopsys</guid>
      <pubDate>Wed, 18 Nov 2026 19:09:42 GMT</pubDate>
      <author>admin@techarchive.dev (Sham)</author>
      <category>AI News</category>
      <description>OpenAI says its models rewrote AMD GPU RTL for 14% PPA gains and silicon-proven Jalapeno savings, as GPT-Synopsys industrializes AI chip design.</description>
    </item>
    <item>
      <title>Xing4: China Telecom Trains 285B Model on Ascend, No Nvidia</title>
      <link>https://aitecharchive.com/articles/xing4-china-telecom-285b-model-ascend-910c-no-nvidia</link>
      <guid isPermaLink="true">https://aitecharchive.com/articles/xing4-china-telecom-285b-model-ascend-910c-no-nvidia</guid>
      <pubDate>Wed, 18 Nov 2026 17:27:27 GMT</pubDate>
      <author>admin@techarchive.dev (Sham)</author>
      <category>World AI &amp; Tech</category>
      <description>China Telecom&apos;s Xingchen AGI Lab trained a 285B MoE on 8,192 Huawei Ascend 910C chips with zero Nvidia silicon. Its open 29B member runs on one GPU.</description>
    </item>
    <item>
      <title>Qwen 27B Rewrites Its Own Weights With No Training</title>
      <link>https://aitecharchive.com/articles/qwen-27b-rewrites-own-weights-no-training</link>
      <guid isPermaLink="true">https://aitecharchive.com/articles/qwen-27b-rewrites-own-weights-no-training</guid>
      <pubDate>Wed, 18 Nov 2026 15:46:39 GMT</pubDate>
      <author>admin@techarchive.dev (Sham)</author>
      <category>AI Research</category>
      <description>A new preprint shows Qwen3.5-27B diagnosing and editing its own weights with no gradients, gaining 6.6 points on held-out problems - and smaller models fail.</description>
    </item>
    <item>
      <title>Apple FastFrames Open Weights: 36 ms Interactive Video</title>
      <link>https://aitecharchive.com/articles/apple-fastframes-open-weights</link>
      <guid isPermaLink="true">https://aitecharchive.com/articles/apple-fastframes-open-weights</guid>
      <pubDate>Wed, 18 Nov 2026 14:07:25 GMT</pubDate>
      <author>admin@techarchive.dev (Sham)</author>
      <category>Open Source</category>
      <description>Apple released FastFrames-S and L video model weights open on Hugging Face on November 16. The 1.3B variant hits 36 ms per frame on one B200 GPU.</description>
    </item>
    <item>
      <title>Razorpay on Meta Business Agent: WhatsApp Sales in One Chat</title>
      <link>https://aitecharchive.com/articles/razorpay-meta-business-agent-whatsapp-sales</link>
      <guid isPermaLink="true">https://aitecharchive.com/articles/razorpay-meta-business-agent-whatsapp-sales</guid>
      <pubDate>Wed, 18 Nov 2026 12:26:19 GMT</pubDate>
      <author>admin@techarchive.dev (Sham)</author>
      <category>AI for Small Business</category>
      <description>Razorpay went live on Meta Business Agent on 17 November 2026. Here is what Indian small businesses can do with in-chat UPI, card and netbanking payments.</description>
    </item>
    <item>
      <title>OpenScientist: Open AI Co-Scientist Passes Expert Tests</title>
      <link>https://aitecharchive.com/articles/openscientist-ai-co-scientist-nature-medicine</link>
      <guid isPermaLink="true">https://aitecharchive.com/articles/openscientist-ai-co-scientist-nature-medicine</guid>
      <pubDate>Wed, 18 Nov 2026 10:46:48 GMT</pubDate>
      <author>admin@techarchive.dev (Sham)</author>
      <category>AI Research</category>
      <description>OpenScientist, an open-source agentic AI co-scientist, passed expert evaluation in Nature Medicine. What it does, what it got right, and how to run it.</description>
    </item>
    <item>
      <title>Karnataka to Train 63,200 in AI With Google and Sarvam</title>
      <link>https://aitecharchive.com/articles/karnataka-ai-skilling-63200-bengaluru-tech-summit-2026</link>
      <guid isPermaLink="true">https://aitecharchive.com/articles/karnataka-ai-skilling-63200-bengaluru-tech-summit-2026</guid>
      <pubDate>Wed, 18 Nov 2026 09:05:20 GMT</pubDate>
      <author>admin@techarchive.dev (Sham)</author>
      <category>India AI &amp; Tech</category>
      <description>Karnataka will train 63,200 students and engineers in AI via MoUs with Google, Salesforce, Pega, Sarvam AI and Gnani.ai, plus six new deeptech initiatives.</description>
    </item>
    <item>
      <title>Copilot Fusion Beats Opus 5.5, Fable 5.2 on Terminal-Bench 4.1</title>
      <link>https://aitecharchive.com/articles/copilot-fusion-tops-terminal-bench-41-beats-opus-55-fable-52-cheaper</link>
      <guid isPermaLink="true">https://aitecharchive.com/articles/copilot-fusion-tops-terminal-bench-41-beats-opus-55-fable-52-cheaper</guid>
      <pubDate>Wed, 18 Nov 2026 07:27:59 GMT</pubDate>
      <author>admin@techarchive.dev (Sham)</author>
      <category>AI Coding Tools</category>
      <description>Fusion Max resolves 71.2% on Terminal-Bench 4.1 — first above 70%. It beats Opus 5.5 and Fable 5.2 by 5.4 points while costing 57% and 88% less.</description>
    </item>
    <item>
      <title>Tencent Hy4-flash Open Weights: 220B MoE, $0.07 per Million</title>
      <link>https://aitecharchive.com/articles/tencent-hy4-flash-base-open-weights</link>
      <guid isPermaLink="true">https://aitecharchive.com/articles/tencent-hy4-flash-base-open-weights</guid>
      <pubDate>Wed, 18 Nov 2026 05:46:40 GMT</pubDate>
      <author>admin@techarchive.dev (Sham)</author>
      <category>AI Models &amp; Benchmarks</category>
      <description>Tencent opened Hy4-flash weights: a 220B MoE with 12B active per token and 1M context, at $0.070 input and $0.281 output per million tokens.</description>
    </item>
    <item>
      <title>SenseNova-Loop-9B Open Weights: 9B Model Ties Qwen 27B</title>
      <link>https://aitecharchive.com/articles/sensenova-loop-9b-open-weights-matches-qwen-27b</link>
      <guid isPermaLink="true">https://aitecharchive.com/articles/sensenova-loop-9b-open-weights-matches-qwen-27b</guid>
      <pubDate>Wed, 18 Nov 2026 04:06:27 GMT</pubDate>
      <author>admin@techarchive.dev (Sham)</author>
      <category>AI Research</category>
      <description>SenseTech AI shipped SenseNova-Loop-9B weights under Apache 2.0 on November 17, 2026. The looped 9B model matches much larger dense rivals on agent benchmarks.</description>
    </item>
    <item>
      <title>Michigan State Data Breach: BYOD Gang Claims 1.3M Records</title>
      <link>https://aitecharchive.com/articles/michigan-state-data-breach-byod-ransomware</link>
      <guid isPermaLink="true">https://aitecharchive.com/articles/michigan-state-data-breach-byod-ransomware</guid>
      <pubDate>Tue, 17 Nov 2026 20:49:11 GMT</pubDate>
      <author>admin@techarchive.dev (Sham)</author>
      <category>Cybersecurity</category>
      <description>MSU confirmed Nov 16, 2026 that contact and demographic data of some students and staff was potentially exposed, after BYOD claimed 1.3M records.</description>
    </item>
    <item>
      <title>GPT-6.2 Astra: OpenAI First Self-Improvement Model</title>
      <link>https://aitecharchive.com/articles/gpt-6-2-astra-launch-price-self-improvement</link>
      <guid isPermaLink="true">https://aitecharchive.com/articles/gpt-6-2-astra-launch-price-self-improvement</guid>
      <pubDate>Tue, 17 Nov 2026 19:09:29 GMT</pubDate>
      <author>admin@techarchive.dev (Sham)</author>
      <category>AI News</category>
      <description>OpenAI shipped GPT-6.2 Astra and Sol on November 17, 2026: frontier scores up, Astra priced 20 percent lower, and a first High self-improvement rating.</description>
    </item>
    <item>
      <title>AI News Today: California Sues OpenAI and 4 More Stories</title>
      <link>https://aitecharchive.com/articles/ai-news-today-november-17-2026</link>
      <guid isPermaLink="true">https://aitecharchive.com/articles/ai-news-today-november-17-2026</guid>
      <pubDate>Tue, 17 Nov 2026 17:28:52 GMT</pubDate>
      <author>admin@techarchive.dev (Sham)</author>
      <category>AI News</category>
      <description>California sues OpenAI over 80 agent incidents; the US licenses China&apos;s remote compute; Google maps no-human exploit machines; Isomorphic Labs raises $5bn.</description>
    </item>
    <item>
      <title>MLPerf v6.1: Vera Rubin vs GB300, Rubin Wins 6 of 7</title>
      <link>https://aitecharchive.com/articles/mlperf-training-v6-1-vera-rubin-vs-gb300</link>
      <guid isPermaLink="true">https://aitecharchive.com/articles/mlperf-training-v6-1-vera-rubin-vs-gb300</guid>
      <pubDate>Tue, 17 Nov 2026 15:49:36 GMT</pubDate>
      <author>admin@techarchive.dev (Sham)</author>
      <category>AI Models &amp; Benchmarks</category>
      <description>NVIDIA&apos;s Vera Rubin NVL72 wins 6 of 7 MLPerf Training v6.1 benchmarks, halving DeepSeek-V3 time per GPU vs GB300, as agent-written kernels threaten CUDA&apos;s edge.</description>
    </item>
    <item>
      <title>Agnes 2.5 Pro Theta Weights: Self-Host vs Cheaper API</title>
      <link>https://aitecharchive.com/articles/agnes-2-5-pro-theta-open-weights-self-host-vs-rent</link>
      <guid isPermaLink="true">https://aitecharchive.com/articles/agnes-2-5-pro-theta-open-weights-self-host-vs-rent</guid>
      <pubDate>Tue, 17 Nov 2026 14:07:33 GMT</pubDate>
      <author>admin@techarchive.dev (Sham)</author>
      <category>Open Source</category>
      <description>Agnes 2.5 Pro Theta went open under Apache 2.0 while its beta-priced hosted API stays cheap. Break-even math, hardware floor, and the pick.</description>
    </item>
    <item>
      <title>Zoho Desk Blended Agents in India: Setup Steps and INR Cost</title>
      <link>https://aitecharchive.com/articles/zoho-desk-blended-agents-india-setup-cost</link>
      <guid isPermaLink="true">https://aitecharchive.com/articles/zoho-desk-blended-agents-india-setup-cost</guid>
      <pubDate>Tue, 17 Nov 2026 12:25:59 GMT</pubDate>
      <author>admin@techarchive.dev (Sham)</author>
      <category>AI for Small Business</category>
      <description>Zoho Desk Blended Agents launched November 16, 2026: AI agents add no extra fee. Setup steps and the INR monthly cost for a small Indian support desk.</description>
    </item>
  </channel>
</rss>