Imodeli enkulu
Ithegi yethuluzi le-AIEZZ
Ubheka imikhiqizo ye-AI enethegi ethi “Imodeli enkulu”, enemiphumela efanayo engu-10.
2026 Ukubuyekezwa Kwesoftware Yokubhala Amanoveli nge-AI: FeelFish 4.0Ukuhlela nge-AII-FeelFish 4.0 iyisoftware yobungcweti yokubhala amanoveli, eguqula amamodeli amakhulu e-AI abe yiqembu lababhali eliqonda iphrojekthi yakho. Ihlinzeka ngokuphathwa kwephrojekthi okuhlelekile, okuhlanganisa izahluko, uhlaka, abalingiswa kanye nezilungiselelo, futhi isekela umongo ohlakaniphile ukuze kugcinwe ukuhambisana komqondo ekubhaleni imisebenzi emide. Leli thuluzi lisekela ukusebenzisana kwama-agent amaningi enziwa ngokwezifiso, ukulawula okukude ngekhompyutha nedivayisi yeselula, kanye nokuvumelanisa nge-cloud drive ne-time machine yendawo ukuze kuvikelwe imibhalo. Lihloselwe ukuhlinzeka ababhali bemisebenzi emide, njengamanoveli ewebhu nemibhalo yemidlalo, ngohlelo oluphelele lokusebenza lokubambisana phakathi komuntu ne-AI.00
RabbitVis – Ithuluzi elihlangene le-AI lokuklama nokukhiqiza elethulwe yi-兔展智能Ukuhlela izithombeI-RabbitVis iyithuluzi eliphelele le-AI lokuklama nokukhiqiza elisebenza ngezendlalelo, elethulwe yi-兔展智能, elisekelwe kumodeli obonakalayo we-UniWorld-Design ozithuthukisele wona. Lihlanganisa ngokujulile amandla okukhiqiza e-AI namandla okuhlela ngezendlalelo. Lisekela ukukhiqiza amaposta, amakhasi emininingwane yemikhiqizo ye-e-commerce, amakhasi angaphakathi omagazini nokunye okuqukethwe komklamo osezingeni eliphezulu ngolimi lwemvelo, kanye nokuhlela okuqhubekayo ngezendlalelo, njengokususa ingemuva, ukushintsha umbhalo, ukufaka ezinye izakhi nokukhipha izinto ezinezizinda ezisobala. Ithuluzi lenzelwe ngokukhethekile izimo zesiShayina, futhi lisekela amafonti obuciko ayinkimbinkimbi, ukuhlanganiswa kombhalo nezithombe kanye namashadi olwazi ekuhlelweni okunobuningi obuphezulu. Lihloselwe abasebenzisi bokusebenza kwe-e-commerce, ukuphathwa kwezokuxhumana, ukukhangisa komkhiqizo, ukudalwa kokuqukethwe kanye nomsebenzi wasehhovisi, libenza bakwazi ukukhiqiza izinto ezibonakalayo ezisezingeni lochwepheshe noma bengenalo ulwazi lokuklama.00
MemHarness – uhlaka lokwakha kabusha inkumbulo oluqalwe yi-Shanghai AI Lab nabanyeI-ejenti ye-AII-MemHarness iwuhlaka lokwakha kabusha inkumbulo lwama-LLM Agent, olwethulwe ngokuhlanganyela yi-Shanghai AI Lab namanyuvesi ahlanganisa i-Zhejiang University, i-Fudan University kanye ne-Shanghai Jiao Tong University. Ifaka izinyathelo zokuhlaziya nokwakha kabusha phakathi kokusesha inkumbulo nokwenza isenzo, igcine, ibhale kabusha noma ilahle okuhlangenwe nakho kwangaphambilini ngokuhambisana nomongo wamanje. Isebenzisa ukuqeqeshwa kwe-GRPO kusukela ekuqaleni kuye ekugcineni ngaphandle kwedatha eyengeziwe enelebula eyenziwe abantu. Uhlaka lusekela yonke inqubo yokusesha inkumbulo, ukuhlaziya nokwakha kabusha, ukukhiqiza izenzo, ukubuyisela okuhlangenwe nakho nokususa inkumbulo enenani eliphansi, futhi lunikeza inqubo yokuthatha izinqumo enezigaba ezinhlanu. Imodeli enamapharamitha angu-7B idlula i-Gemini-2.5-Pro ngo-23.1% ku-ALFWorld nango-39.7% ku-WebShop, kanti izinga layo lokuphumelela elijwayelekile ezimweni ezingaphandle kokusatshalaliswa lifinyelela ku-85.9%. Isebenziseka emisebenzini yasekhaya yamarobhothi ahlakaniphile, ekuthengeni ku-inthanethi ngokuzenzakalelayo, ekusekelweni kwamakhasimende okuhlakaniphile, ekusizeni ukuthuthukiswa kwekhodi nakwezinye izimo.10
dots.ttsUkwenziwa kwenkulumo nge-AII-dots.tts iyimodeli eyisisekelo yokwenziwa kwenkulumo ye-autoregressive eqhubekayo ngokuphelele, enamapharamitha ayizigidigidi ezi-2, ekhishwe ngomthombo ovulekile ngokubambisana yiqembu le-Xiaohongshu dots kanye ne-X-LANCE Laboratory yase-Shanghai Jiao Tong University. Imodeli ikhiqiza umsindo we-48kHz ngamaqoqo ngokuqondile endaweni efihlekile eqhubekayo, futhi...00
SeedRealtime – Imodeli yomdabu ye-full-duplex yomsindo nevidiyo eyethulwe yi-ByteDanceAmamodeli amakhulu e-AII-SeedRealtime – imodeli yomdabu ye-full-duplex yomsindo nevidiyo eyethulwe yi-ByteDance, iyimodeli yomdabu ye-full-duplex yomsindo nevidiyo eyethulwe ithimba le-ByteDance Seed. Ihlanganisa umsindo, ividiyo nombhalo ngokwemvelo ngaphansi kwesakhiwo esihlanganisiwe, ukuze ikwazi ukuxhumana kwesikhathi sangempela kwe-multimodal ngokubuka, ukulalela nokukhuluma ngesikhathi esisodwa. Imisebenzi yayo esemqoka ihlanganisa ukuqonda okuhlangene komsindo nevidiyo, ikhono lokuxhumana ngokuqhubekayo, ukuhamba kwengxoxo okushelelayo kanye nokubona izigcawu ezinabantu abaningi. Ingasebenzisa isimo esibonakalayo ukuqeda ukungaqondakali kwamagama aphinyiselwa ngokufanayo, ikhumbuze abasebenzisi ngokuqhubekayo lapho isimo esikrinini sishintsha, ihlukanise izingxoxo zabanye abantu nomsindo wangemuva ezindaweni ezinomsindo, futhi iqhubeke ihlobanisa izwi lomuntu ngamunye nobunikazi bakhe. Imodeli isebenzisa ukumodela okuhlanganisiwe kusukela ekuqaleni kuze kube sekugcineni kanye nendlela yomdabu yokuxhumana kwe-full-duplex, ngaleyo ndlela igweme ukulahleka kolwazi nokunqwabelana kokubambezeleka ezinhlelweni ezixhunywe ngezigaba. Isivele yethulwe kubo bonke abasebenzisi bohlelo lwe-Doubao App, yaba umkhiqizo wokuqala we-AI womsindo nevidiyo we-full-duplex osetshenziswe ngezinga elikhulu embonini.10
HOMIEUkukhiqiza ividiyo nge-AII-HOMIE iwuhlaka oluvulekile lokukhiqiza amavidiyo abantu bedijithali olukhishwe yi-Hong Kong University of Science and Technology, olwakhelwe kunethiwekhi eyinhloko i-Wan2.1-T2V-14B futhi oluhlanganisa imodeli enkulu yezindlela eziningi i-Qwen3-VL. Uhlaka luphatha ngokuhlangene izici ezine: umuntu wedijithali + umkhiqizo + i-Logo + umbhalo, ...00
SenseNova U1.5-Lite-Preview – Imodeli ye-multimodal elula, evulekile yakwa-SenseTimeAmamodeli amakhulu e-AISenseNova U1.5-Lite-Preview – imodeli ye-multimodal elula, evulekile yakwa-SenseTime, isekelwe ekwakhiweni kwe-NEO-Unify futhi isebenzisa amapharamitha angu-8B-MoT ukuze ihlanganise ukuqonda okubukwayo, ukucabanga, ukwakha nokuhlela ngendlela yemvelo. Imodeli isekela ukwakha izithombe ze-ultra-HD ze-4K ngendlela yemvelo, inikeza ukuthungwa okunemininingwane, ukubukeka okungokoqobo kanye nekhono lokunikeza umbhalo wesiShayina nesiNgisi ezakhiweni eziyinkimbinkimbi. I-Prompt Enhance Skill eyakhelwe ngaphakathi yehlisa umngcele wokudala, ikwazi ukusebenzisa ngokunembile imiyalelo emide kakhulu yolimi lwemvelo kanye nemiyalelo ebonakalayo ehlelekile, futhi isekele ukuhlela okuqhubekayo nokuphindaphindwayo. Iqondiswe kubaklami, abathuthukisi nabadali bokuqukethwe, inikeza ithuluzi lokudala okubukwayo elisezingeni lezentengiso, elinezindleko eziphansi zokufakwa kanye nezinga eliphezulu lokulandela imiyalelo.10
IndexTTS-2.5Ukwenziwa kwenkulumo nge-AII-IndexTTS-2.5 iyimodeli yokukopisha izwi esezingeni lezimboni ye-zero-shot, ekhishwe yi-Bilibili njengomthombo ovulekile. Inamapharamitha angu-0.8B kuphela, futhi isekela ukudluliselwa kwezwi phakathi kwezilimi ezinhlanu: isiShayina, isiNgisi, isiJapane, iSpanishi nesi-Arabhu.00
出门问问Inkundla yokufunda nge-AI出门问问 yinkampani ehlinzeka ngemikhiqizo nezixazululo zobuhlakani bokwenziwa, ehlanganisa ukwenziwa kokuqukethwe, ukuxhumana ngezwi, ukuguqula okuqoshiwe kube umbhalo kanye nezingxenyekazi zekhompuyutha ezihlakaniphile. Abasebenzisi bangakhetha amathuluzi okudubha ngezwi kusetshenziswa ubuhlakani bokwenziwa, okukhiqiza abantu bedijithali noma okudala amavidiyo kuye ngomsebenzi. Bangasebenzisa ne-TicNote ukuqopha, ukuguqula inkulumo ibe umbhalo ngesikhathi sangempela, ukwenza izifinyezo nokubuza imibuzo. Lezi zinsizakalo zilungele abadali bamavidiyo amafushane, abasebenzi bezokumaketha nabezindaba ezizimele, ochwepheshe abadinga ukuhlela izinto zemihlangano, kanye nezinkampani ezihlola izinsizakalo zedijithali. Inkundla inikeza nemikhiqizo yamahhala nekhohliwe; imisebenzi nendlela yokuyisebenzisa kufanele kuqondwe ngokwehlukana ngomkhiqizo ngamunye. Okuqukethwe okukhiqiziwe nemiphumela yokuguqula inkulumo ibe umbhalo kufanele kuhlolwe umuntu.018
Qwen 3.8-Max – imodeli yolimi ehamba phambili eyethulwe yiQwen Team yakwa-AliyunAmamodeli amakhulu e-AII-Qwen 3.8-Max – imodeli yolimi ehamba phambili eyethulwe yiQwen Team yakwa-Aliyun, eyakhiwe ngokwandisa i-architecture ye-Qwen 3.5 yafinyelela kumapharamitha angu-2.4 trillion, futhi ihloselwe ukuba imodeli yolimi esezingeni elikhulu kakhulu yamabhizinisi nocwaningo. Isekela ukubhala ikhodi okuzenzakalelayo isikhathi eside, ukuphinda izivivinyo zamaphepha ocwaningo, imincintiswano, ukuthuthukisa ukusebenza kwehhovisi, ukuthuthukiswa kwamasu e-quantitative kanye nokuziklanyela kwama-chip. Imodeli isebenzisa i-reinforcement learning yomhlaba wangempela kanye ne-feedback loop yokuzithuthukisa ukuze ikwazi ukusebenza ngokuzenzakalelayo izinsuku ezimbalwa kuya emashumini ezinsuku, iqede imisebenzi eyinkimbinkimbi ngaphandle kokungenelela komuntu. Abasebenzisi bangayibiza nge-API ehambisana ne-OpenAI noma i-Anthropic protocol esesiteji se-Qianwen AI, noma basebenzise ama-open-source weights endaweni, ukuze bayihlanganise kalula nama-agent frameworks afana ne-Claude Code, Codex kanye ne-Qoder. Ngenxa yalokho, amaqembu ocwaningo nentuthuko, ochwepheshe bezomthetho nokuhambisana nemithetho, abaklami nabanye abasebenzisi abangochwepheshe bangakhuphula kakhulu ukusebenza kwabo, banciphise amaphrojekthi athatha isikhathi eside abe ukulethwa ngaphakathi kwengxoxo eyodwa.10Wonke amathuluzi analeli thegi ayaboniswa.