Status
SteamPeaks
ChartsSalesUpcomingPatchesNewsCalculator
New on SteamEvery app, DLC and depot the minute Steam creates itAppsEvery app on Steam, newest change firstPackagesSubs and bundles, and what each containsDepotsDepots, manifests and install sizesTagsSteam's user tags and the games under themDevelopers & publishersCompanies and their cataloguesTechnologiesEngines, SDKs and anti-cheat found in the filesChange historyEvery PICS changelist as it lands
SignalsNineteen readings of the whole catalogueCompareAny games side by sideRecordsAll-time peaks and the days they were setReportsWeekly and monthly write-upsAlerts & newsroomWatch a game, get told when it movesSteam statusIs Steam up, right nowWeb API explorerTry the endpoints in the browser
NewsSteam's own announcements and the sales calendarCalculatorWhat a Steam account is worth, and its pile of shame
/
Sign in
/
SteamPeaks
The ultimate resource for Steam data.
ExploreChartsSalesSales and festsUpcomingPatchesNewsRecordsTrendingSignals
DatabaseAppsPackagesDepotsTagsDevelopersTechnologiesChange history
ToolsCalculatorCompareSearchAlertsSteam statusAPI
SiteMethodologyFAQDiscordSupportSign in via Steam
Not affiliated with Valve or Steam. Game names and artwork belong to their owners. All times UTC.
PrivacyCookiesFair useStatus
SelbyNewsSelby Update: Complete Answers, Smarter Analysis, Auto Language
Game update

Selby Update: Complete Answers, Smarter Analysis, Auto Language

Selby · published 27 Aug 2026, 13:59 UTC

All newsPlayers around this dateRead on Steam

What's new in this update:

1. Complete answers, always
Long replies were sometimes cut off mid-sentence. The reply length limit has been doubled — and if an answer is still longer, Selby now automatically continues from where it stopped. You get the full reply in one piece.

2. Project analysis no longer gives up
Analyzing large files or multiple projects used to overwhelm the AI and make it quit early. Selby now reads large files in sections, notes the key points of each file as it goes, and keeps working from those notes — it walks through your project step by step instead of saying "impossible".

3. The right language from the first launch
Selby now detects your system language on first launch. Most of the world will see an English interface by default — and you can switch to any of the 10 supported languages anytime in Settings.

4. A clearer explanation of how the AI works
We've rewritten the store description to be upfront: the built-in AI runs on our cloud servers and needs an internet connection, and the one-time starter token quota included in your purchase never expires.

Special thanks to player FastRabbit for the detailed feedback — this update is built around it. Every review and report shapes the next update.

More from Selby

Other announcements

All news
Community27 Aug 2026Selby: Bug Report & Feedback ThreadThank you for buying and supporting Selby! This post is the official place to report bugs — drop any issue you run into in the comments below. Version 3.22.1 just went live with: drag & drop files (and folders) into the chat window to copy their paths, multi-line input, skill installation from the Steam Workshop, and a welcome achievement. To help me fix things faster, please include: 1. Your W…Game update26 Aug 2026Selby Update: Drag & Drop Paths, Multi-line Input, Workshop SkillsWhat's new in this update: 1. Drag files into the chat window Drag one or more files from Explorer onto the Selby window — their full paths are copied to the clipboard and inserted into the input box. Send them right away or paste them anywhere. 2. Multi-line input The input box now grows with your text (up to 6 lines). Enter sends, Shift+Enter adds a new line — long prompts are no longer squee…Game update23 Aug 2026Selby Update: Token Efficiency Fix + Token PacksThanks to player "좋아하는게임을반영한닉네임을사용하세요" for the feedback. The token consumption issue you reported has been identified and fixed. What's new: 1. Token efficiency fix Timestamps were previously injected into the system prompt on every message, defeating the model's context caching and inflating input costs. With caching working now, cached input costs about 1/30 of before — the same quota goes mu…