Xiga xigaHagaha xiga
Qaabka Guud ee Hab-maamuuska
Luuqadda AI
HAGAHA Shirkadaha
Magic AI Moodooyinka Koodhka dheer ee macnaha guud
Magic AI is a startup developing long-context models for software-development tasks.
Dhageyso hagahanAI-waa laga sheekeeyay · ~2 min
Boggaan2 daqiiqo akhri
Dulmar
The company says LTM-2-mini supports up to 100 million tokens, which can place substantially more cross-file code and documentation in one context; capacity alone does not guarantee correct reasoning over an entire codebase.
quusid qoto dheer
Magic AI is a startup developing an AI software engineer rather than only autocomplete. In its August 2024 research update, the company described LTM models, including LTM-2-mini, with a context window of up to 100 million tokens; its analogy is roughly 10 million lines of code or about 750 novels. This is a model capacity claim, not evidence that the model can correctly understand every codebase or complete every project-wide task. The update also describes a Google Cloud partnership for supercomputer development and funding announced at that time. Magic introduced HashHop as a controlled evaluation using random, incompressible hash pairs and chained lookups to test retrieval at long context lengths. Results on that constructed task measure those retrieval and chaining tasks; they do not prove arbitrary code comprehension or performance on every real repository.
Dhaqangelinta Adduunka-dhabta ah
Providing a large repository or selected subset as context for questions about cross-file relationships, then checking the answer against source code.
Drafting a refactor that touches related files for a developer to review and test.
Supplying code, documentation, and logs together when investigating a bug whose cause may span files.
Summarizing a selected repository context as an onboarding aid, with links or references for follow-up.
Mustaqbalka Magic AI Moodooyinka Koodhka Dheer ee Macnaha Guud
Longer contexts can allow developers to supply more code, documentation, and task details together. Reliable multi-file work still depends on retrieval, reasoning accuracy, tool execution, and verification. Future systems should be evaluated on representative codebases and tasks, including whether they find relevant distant files and make changes that pass tests, rather than being judged by context size alone.
Ku dhaqangelinta
Maxay muhiim u tahay
- Istaraatiijiyadda iibiyaha. Wadooyinka iibiyuhu waxay saameeyaan sifooyinka ay kooxdaadu dhisi karto marka xigta.
- Qiimaha iyo miisaaniyada. Shuruudaha ganacsiga iyo ikhtiyaarka geynta ayaa saameeya kharashka iyo khatarta muddada-dheer.
- Khatarta iyo badbaadada. Dhiirrigelinta shirkadu waxay qaabaysaa wax-soo-saarka alaabada, booska badbaadada, iyo furfurnaanta.
Iska ilaali
- Ogeysiisyada bilawga ah ayaa laga yaabaa inay ka dheereeyaan xasilloonida socodka shaqo ee wax soo saarka dhabta ah.
- Qiimaynta API ama isbedelada siyaasada waxay jebin karaan malo awaalka habeen dhaxa
- Ku-tiirsanaanta iibiyuhu wuxuu kordhiyaa qufulka iyo kharashaadka socdaalka.
Tallaabooyinka xiga
- Qiimee bixiyeyaasha adiga oo isticmaalaya hawlahaaga iyo kaydinta xogtaada.
- Dib u eeg sirta, amniga, iyo shuruudaha sharciga ka hor is dhexgalka.
- Ilaali qorshaha dib u dhaca ee moodooyinka ama iibiyeyaasha.
- La soco qoraalada sii daynta si isbeddellada khariidadda waddo aysan ula yaabin kooxaha.
Sii wad Sahaminta
Warqad bilaash ah
Hel warbixinta maalinlaha ah ee AI
Saddex sheeko oo la xaqiijiyay oo AI ah maalin kasta oo shaqo ah, oo lagu qoray Ingiriis fudud. Bilaash had iyo jeer, aan xayaysiis lahayn.
Hal iimayl maalin kasta oo shaqo ah. Ka bax hal gujin. Marna ma iibinno ama ma wadaagno ciwaankaaga.
Is tijaabi
Take the Magic AI Long-Context Code Models quiz
Fikrad degdeg ah oo ku saabsan jawaab kasta, iyo shahaado la wadaajin karo oo leh aqoonsi la xaqiijin karo markaad ka gudubto koorso.
Taageer waxbarashada AI bilaashka ah. AI Understanding waa 501(c)(3) aan faa'iido doon ahayn — xayaysiis la'aan, albaab lacag ah ma jiro, had iyo jeer. Samee deeq
Su'aalaha soo noqnoqda
What is Magic AI Long-Context Code Models?
Magic AI is a startup developing long-context models for software-development tasks. The company says LTM-2-mini supports up to 100 million tokens, which can place substantially more cross-file code and documentation in one context; capacity alone does not guarantee correct reasoning over an entire codebase.
Waa maxay jihada horumarinta software-ka uu Magic sheegay in uu ku raad joogo?
Magic wuxuu qeexayaa shaqadeeda sida horumarinta AI ee injineernimada software. Tani waa yoolka shirkadda; ma dammaanad qaadayso in moodalku si sax ah u maamuli karo codebase kasta.
Waa maxay awoodda macnaha guud ee Magic ku dhawaaqay LTM-2-mini Agoosto 2024?
Warbixinta Magic's Agoosto 2024 ayaa ku dhawaaqday daaqad macno leh oo ilaa 100 milyan ah oo calaamad ah LTM-2-mini. Awoodda macnaha guud way ka duwan tahay saxnaanta hawsha.
Muxuu Magic u isticmaalaa HashHop si uu u qiimeeyo?
Magic waxa uu HashHop u qaabeeyay sidii xashiish random-ku-soo-celin ah oo la xakameeyey iyo calaamadaynta silsiladda. Waxay yaraynaysaa tibaaxaha micnaha laakiin ma dejiso fahamka kood ballaadhan.
Waa maxay sababta fiiro gaar ah u yeelashada transformer-ka ay dhibaato ugu tahay 100-milyan ee macnaha guud?
Qiimaha dareenka caadiga ah wuxuu ku koraa labajibbaaran dhererka isku xigxiga, sidaa darteed xaaladaha aadka u dheer aad bay qaali u yihiin iyada oo aan lahayn algorithms ka wanaagsan.
Shirkaddee la wadaagtay Magic 2024 si ay u dhisto kombuyuutarrada waaweyn?
Magic wuxuu ku dhawaaqay iskaashi lala yeesho Google Cloud si loo dhiso kombuyuutarrada sare ee qalabka Nvidia.
Sii wad waxbarashada
Tilmaamaha la xidhiidha
Tilmaamayaal badan ayaa loo doortay mawduucan