source&pool
A daily wire of long-form journalism, video, and discourse — filed, tagged, and laid out flat.
VOL. I·NO. 01
WEDNESDAY, SEPTEMBER 16, 2026
Hacker News3775X 主题热门3655MacRumors79CNBC77YahooFinance649to5Mac62Kotaku44Verge43IGN349to5Google32Gematsu32aihot31NintendoLife31Engadget26TechCrunch26Eurogamer25BusinessInsider23Guardian20NBC16CNET15NPR15FoxBusiness14Polygon14Fortune13SeekingAlpha13bgr12Gizmodo12CBS11Mashable11Wccftech11Investor'sBusinessDaily10PushSquare10TechPowerUp10USAToday10WIRED10CNN8NintendoEverything8Notebookcheck8NewYorkPost8CrudeOilPricesToday8VideoGamesChronicle8ABC7ArsTechnica7Fox7GameInformer7WindowsCentral7BleepingComputer6PetaPixel6AppleInsider5Deadline5DigitalFoundry5GamesIndustry.biz5Variety5Yahoo5AlJazeera4AndroidPolice4DroidLife4MotleyFool4GameRant4Jalopnik4PureXbox4SamMobile4SlashGear4Hacker4AP3CTech3CanonRumors3ChromeUnboxed3CoinDesk3GameDeveloper3GSMArena3Motor13Blizzard3XBOXWire3PCMag3PCWorld3SeattleTimes3Space3Register3TweakTown3YGOrganization3ZDNET324/7WallSt.2Aftermath2AndroidCentral2AwfulAnnouncing2BleedingCool2BuzzFeed2DigitalCameraWorld2DualShockers2DW2EventHubs2Futurism2GearPatrol2Hodinkee2Independent2Lifehacker2MassivelyOverpowered2MyNintendo2Nature2Newser2Newsweek2PaulKrugman2PokémonGOHub2RoadtoVR2RockPaperShotgun2RPGSite2Conversation2Intercept2NextWeb2Tom'sGuide2UploadVR2VideoCardz2WarhammerCommunity2WindowsLatest2YourTango2404Media143rumors1ABC111AboveLaw1ageofempires1AndroidHeadlines1AOL1AVClub1Benzinga1BikeRadar1Billboard1BloodyDisgusting1Borderlands1Bungie1Yahoo!FinanceCanada1CineD1CnEVPost1comicbook1CreativeBloq1CyberSecurityNews1DailyKos1DCRainmaker1derekthompson1Draftsim1CNN1Euronews1flatpanelshd1FratelloWatches1FrequentMiler1GAMINGbible1garymarcus.substack1GeekWire1GeekyGadgets1Hackaday1HollywoodReporter1InsiderGaming1InterconnectsAI1InterestingEngineering1JapanTimes1KITCO1KrebsonSecurity1KSL1LosAngelesTimes1Lloyd'sList1WPLGLocal101Macworld1Maxroll1Mediaite1MiddleEastEye1MonochromeWatches1MPR1SemiAnalysis1Newsshooter1NoMan'sSky1nylon.com.sg1NYT1OregonLive1PCGamesN1PersonaCentral1Pokemon1politico.eu1PittsburghPost-Gazette1QuantaMagazine1qz1SammyGuru1ScienceAlert1ScientificAmerican1SouthChinaMorningPost1Semafor1SFGATE1YahooFinanceSingapore1YahooSingapore1SportsIllustrated1SimpleFlying1Sources1supercarblondie1Tedium1TelecomTalk1GameBusiness1TheGamer1Times1LongmontTimes-Call1TmoNews1TopGear1TwistedVoxel1YahooFinanceUK1UnHerd1vox1WhatHi-Fi?1WPBF1WRAL1
  1. 001Hacker NewsSEP · 16English

    GPT-6 Astra vs. GPT-5.6 Sol: Is a 1.6x Higher Cost Worth It per Verified Bug?

    A comparison of GPT-6 Astra and GPT-5.6 Sol code review models found that despite Astra's 2.5x higher cost and superior precision (95% vs 85%), Sol identified more confirmed bugs across 50 pull requests (107 vs 91) at lower cost per bug ($0.039 vs $0.062). The study highlights that meaningful code review evaluation requires measuring both bug detection and false-positive rates, not just raw findings.

    By Aditya Jha
  2. 002Hacker NewsSEP · 15English

    Show HN: Analyzed 400 routes and driving a Tesla beats flying 65% of the time

    Analysis of 430 real trips comparing Tesla driving versus flying shows driving wins 64% of the time overall, but flying becomes cheaper beyond approximately 350 miles. Driving costs $0.12–$0.39 per mile depending on distance, while flying ranges from $0.34–$1.74 per mile, and electric vehicles produce 7.4 times less CO2 than planes on 427 of 430 trips analyzed.

    By Fly or Tesla
  3. 003Hacker NewsSEP · 15English

    Show HN: Sunk Cost – How long until a local LLM rig pays for itself?

    A tool calculating break-even timeframes for local LLM hardware investments by comparing costs against API usage savings. Users can adjust assumptions about speed, memory, and usage patterns to determine when owning local inference infrastructure becomes cost-effective.

    By rlindsey123
  4. 004Hacker NewsSEP · 14English

    Show HN: Replay Doctor – Local CLI to find silent prompt cache leaks

    Replay Doctor is a local CLI tool that identifies prompt cache failures in Claude API usage by analyzing transcripts and pinpointing which conversation turn caused the cache to break, what caused it, and how many tokens were re-billed at write prices. The tool is free, requires minimal setup, and can optionally contribute anonymized diagnostic data to a public pool for aggregated analysis.

    By danielsaito
  5. 005Hacker NewsSEP · 11English

    RTK reports token savings, but our cost benchmarks disagree

    RTK, a popular terminal output compression tool with 79k GitHub stars, claims to reduce AI coding token costs by up to 60–90%, but benchmarking by JetBrains's SkillsBench and independent testing on Terminal-Bench 2.1 found minimal or negative savings. Testing across 1,740 attempts showed costs fell 5% for Fable but rose 5% for DeepSeek; RTK's reported token savings miscount removed output rather than actual billing impact and can mask increased costs from additional model turns.

    By Bartosz Kotrys; Jacek Migdal