Wonke amathuluzi e-AI
Ibhokisi lamathuluzi e-AI
Ubheka imikhiqizo ehlakaniphile ye-AI edumile emkhakheni wamabhokisi amathuluzi e-AI, eyakhelwe wena ngokuthuthukiswa kokufanisa futhi ehlinzeka imboni ngezixazululo ezivelele.
VeriSpoofAmamodeli amakhulu e-AII-VeriSpoof iyimodeli enkulu yomsindo enomthombo ovulekile eyakhiwe yiNyuvesi yaseMelbourne, enamapharamitha angu-2.85B kodwa yaqeqeshwa ngamasampuli angu-29.7M kuphela. I-VeriSpoof ihlanganisa uphenyo lokukhohliswa kwezwi lube ukuphendula imibuzo yemodeli; izisindo ezifanayo zenza ukutholwa kobuqiniso, ukulandelela uhlelo, ukuqinisekisa umsuka ofanayo,...00
Ling-3.2-flashAmamodeli amakhulu e-AII-Ling-3.2-flash iyimodeli enkulu yakamuva yemultimodal yomdabu eyethulwe yi-Ant Ling, esebenzisa ukwakheka kwe-MoE, enamapharamitha angu-560B isiyonke futhi ivule cishe angu-25B kuphela esinyathelweni ngasinye. Iwindi lomongo lifinyelela ku-1M, futhi isekela ngokomdabu okokufaka kwezithombe, umbhalo namavidiyo.00
Wenxin 5.5 Flash BaseAmamodeli amakhulu e-AII-Wenxin 5.5 Flash Base iyimodeli yakamuva eyisisekelo ekhishwe yi-Baidu njengomthombo ovulekile. Isebenzisa ukwakheka kwe-sparse MoE, inamapharamitha ayizigidi eziyizinkulungwane ezingama-280 esewonke, ivuse amaphamethara ayizigidi eziyizinkulungwane eziyi-14 kuphela ngethokheni ngayinye, yaqeqeshwa kusengaphambili ngamathokheni ayizigidigidi ezingama-33.5, futhi isekela umongo wamathokheni ayisigidi esi-1.00
GPT-6 CyberInkundla yokuthuthukisaI-GPT-6 Cyber iyimodeli ye-OpenAI ekhethekile kwezokuphepha kwamasistimu e-computer, esekelwe ku-GPT-6 Astra futhi yaqeqeshwa ngokufunda okuqiniswayo kwemisebenzi yezokuphepha. Ngaphansi kohlaka lwe-Preparedness ithole isilinganiso esiphezulu se-Critical kwezokuphepha kwamasistimu e-computer, futhi itholakala kuphela kumakhasimende e-Daybreak Red agunyaziwe ngemva kokubuyekezwa...00
WeVisOne-VLUmphakathi we-AI yomthombo ovulekileI-WeVisOne-VL iyimodeli eyisisekelo yokubona nolimi ekhishwe njengomthombo ovulekile yiqembu le-Tencent WeChat Vision, esekelwe ku-Qwen3-VL, futhi itholakala ngezinguqulo ezimbili: i-2B ne-4B. I-WeVisOne-VL isebenzisa indlela ebumbene yokubikezela ama-token ahlelekile ukuze imodeli eyodwa...00
EvaOmniUkuhlolwa kwamamodeli e-AII-EvaOmni iyibhentshimakhi yokuhlola amamodeli amakhulu anezindlela zonke, ekhishwe njenge-open source yi-Shanghai Jiao Tong University, Fudan University, Tencent YouTu nabanye. Iqukethe amavidiyo angempela angu-392, imibuzo nezimpendulo engu-1,027 enelebuli ngabantu, ihlanganise amakhono angu-15 namazinga okucabanga angu-3. I-EvaOmni idinga...00
FlashProveInkundla yokuthuthukisaI-FlashProve iwuhlelo lwe-Anthropic oluvulekile lokuqinisekisa ukucabanga, olusebenzisa ubufakazi bolwazi oluyimfihlo ukuze abahlinzeki bezinsizakalo zamamodeli bafakazele ukuthi impendulo ethile ikhiqizwe imodeli ethile ngaphandle kokudalula izisindo zayo.00
MOSS-LiveUkuhlolwa kwamamodeli e-AII-MOSS-Live iwuhlelo lokuxhumana kwesikhathi sangempela olugcwele lwe-duplex olwethulwe yi-Mosi Intelligent. I-MOSS-Live yenza i-AI ilalele futhi ikhulume ngesikhathi esisodwa njengomuntu wangempela, inqume ngokwezinga lencazelo ukuthi iphendule nini, ilinde nini noma yamukele ukuphazanyiswa, ngaphandle kokucushwa yimisindo yokuvuma noma imisindo yasemvelweni; isekela ividiyo...00
云策OmniInkundla yokuthuthukisa云策Omni iyinkundla yomsebenzi ye-AI yasendaweni yedeskithophu esebenza ezinkundleni eziningi, futhi ingumkhiqizo wokuqala we-inthanethi ovumela abasebenzisi ukwenza ngokwezifiso amakhono e-AI. 云策Omni ihlanganisa “amamodeli＋amakhono＋ama-plugin＋izindawo zokusebenza ezibonakalayo”, ukuze i-AI ikwazi ukuqedela...00
android-ad-cleanerUmphakathi we-AI yomthombo ovulekileandroid-ad-cleaner (Ukususwa kwezikhangiso ku-Android) iyithuluzi lase-China elinomthombo ovulekile lokususa izikhangiso kumafoni e-Android ngokuchofoza okukodwa. I-android-ad-cleaner ixhuma ifoni nge-ADB, ivale ngokuzenzakalelayo izikhangiso ezifihlwe ngokujulile, usesho oludumile...00
SenseNova Loop-9B-PreviewAmamodeli amakhulu e-AII-SenseNova Loop-9B-Preview iyimodeli yokubuka kuqala ye-multimodal enezinga lamapharamitha angu-9B, ekhishwe yi-SenseTime njengomthombo ovulekile. I-SenseNova Loop-9B-Preview igcina izendlalelo zokudekhowuda zoqobo ezi-4 ekuqaleni nasekupheleni, kanti izendlalelo ezimaphakathi ziguqulwa zibe amamojuli okubala aphindaphindwayo abelana ngamapharamitha, ...00
Nemotron 3.5 TranscribeAmamodeli amakhulu e-AII-Nemotron 3.5 Transcribe iyimodeli enkulu evulekile yenkulumo eyethulwe yi-NVIDIA, esebenzisa ukwakheka kwe-encoder yenkulumo ne-MoE, enamapharamitha asebenzayo angaba yizigidi eziyizinkulungwane ezingu-4.3, isekela izilimi ezingu-40, futhi ingacubungula okurekhodiwe okufika ehoreni elingu-1 okunabantu abafika kwabangu-8 abakhulumayo...00
I-Radar Yamagama AmashaUmphakathi we-AI yomthombo ovulekileI-Radar Yamagama Amasha (new-term-radar) iyithuluzi elivulekile lokuqapha amagama amasha e-AI, elisiza abasebenzisi bathole amagama amasha ngaphambi kokuba afakwe kumathuluzi angukhiye. I-Radar Yamagama Amasha iskena ngokuzenzakalelayo ama-akhawunti e-X, uhlu oludumile lwe-Hugging Face, izinqolobane ze-GitHub eziqhuma...00
Hy Memory 2.0Inkundla yokuthuthukisaI-Hy Memory 2.0 iwuhlaka lwememori lwama-agent olukhishwe yi-Tencent Hunyuan njengomthombo ovulekile, olusebenzisa ukwakheka kwezingqimba ezintathu: imodeli + uhlaka + uhlelo lokusebenza: imodeli i-Hy-Memory-30A3B eqeqeshelwe inkumbulo, i-SDK esekela izinkumbulo zengxoxo/umsebenzi/ikhodi, kanye...00
MiDashengLM-SpatialAmamodeli amakhulu e-AII-MiDashengLM-Spatial iyimodeli enkulu yokuqonda umsindo wendawo evulelwe umphakathi yithimba lakwa-Xiaomi Dasheng, futhi iyimodeli yokuqala embonini ehlanganisa ukuqonda umsindo okuvamile nokuqonda indawo ngaphakathi kwesakhiwo esisodwa, kusukela ekuqaleni kuze kube sekugcineni.00
Muse Spark OpenAmamodeli amakhulu e-AII-Muse Spark Open iyimodeli enkulu ye-multimodal evulekile ye-Meta, esebenzisa ukwakheka kwe-sparse mixture-of-experts, enamapharamitha angu-8160 billion kanye namapharamitha angu-42 billion asebenzayo ekuqageleni ngakunye, isekela umongo ofika ku-1 million Token futhi ingacubungula umbhalo, izithombe,...00
DMADUmphakathi we-AI yomthombo ovulekileI-DMAD (Distribution Matching as Adversarial Distillation) iwuhlaka lokucindezela olunezinyathelo ezimbalwa, olukhishwe njengomthombo ovulekile ngokubambisana yi-ByteDance ne-Texas A&M University, olusheshisa ukukhiqizwa ngamamodeli okusabalalisa. I-DMAD iqeqesha umhlukanisi wezigaba ezimbili...00
MiniCPM-V 4.7Amamodeli amakhulu e-AII-MiniCPM-V 4.7 iyimodeli enkulu ye-multimodal esebenza kumadivayisi, ekhishwe njenge-open source ngokubambisana yi-Mianbi Intelligence ne-OpenBMB, futhi inezinguqulo ezimbili: i-35B-A3B (MoE) ne-1B.00
Muse WonderInkundla yokuthuthukisaI-Muse Wonder iyimodeli esezingeni eliphezulu eyethulwe yi-Meta Superintelligence Labs, eyakhelwe phezu kwe-Muse Spark, enenkumbulo yesikhathi eside phakathi kwezingxoxo kanye nekhono lokuqhubeka ngokuzimela nemisebenzi emide; uma ibhekana nesithiyo, ishintsha indlela ngokwayo, futhi iqinisekisa nomsebenzisi ngaphambi kwezenzo ezibalulekile...00
EveryInfraInkundla yokuthuthukisaEveryInfra (Injini Yomhlaba Wonke) iyinkundla yengqalasizinda yezinhlelo zokusebenza ze-AI, ehlinzeka ngamakhono amathathu amakhulu: ukuqoqwa kwedatha ezinkundleni, ukusesha kwe-AI okuxhunywe ku-inthanethi, nokubonwa kwama-CAPTCHA. Ulayini wemikhiqizo we-EveryData uhlanganisa cishe izinkundla ezingu-90 nezindawo zokuxhumana zedatha ezingu-409...00
ClefAmamodeli amakhulu e-AII-Clef iwumndeni wamamodeli ezinqumo ovulekile we-Cloudflare, obeka umongo kuqala bese unquma isinyathelo esilandelayo. Ngokungafani namamodeli ezingxoxo akhiqiza umbhalo ovulekile, i-Clef ichaza umbuzo okwesikhashana esicelweni, inikeze inketho ngayinye ...00
Microsoft-Decision-1Ukuhlolwa kwamamodeli e-AII-Microsoft-Decision-1 iyimodeli ye-AI yesizukulwane esisha yokwenza izinqumo ngokushesha eyethulwe yi-Microsoft. Yakhelwe ngokukhethekile imisebenzi yezinqumo ehlelekile, ikhetha isisombululo esingcono kakhulu ezinkethweni ezichazwe kusengaphambili futhi inikeze impendulo ngayinye amaphuzu wamathuba ukuze izinhlelo ezilandelayo zinqume ukuthi ziqhubeke, zizame futhi noma zidlulisele kumuntu...00
VoxMemAmamodeli amakhulu e-AII-VoxMem iyibhentshimakhi yememori yamamodeli amakhulu omsindo eyethulwe amaqembu avela eNyuvesi yaseMelbourne naseNyuvesi yaseNew South Wales. Iqala ukusebenzisa izinhlobo ezi-4 zobufakazi be-acoustic: incazelo, isikhulumi, ithoni, nomsindo wendawo × imisebenzi emi-4 yememori: ukukhipha, ukucabanga phakathi kwamaseshini, ukulandelela ngokulandelana kwesikhathi...00
Mistral Large 4Amamodeli amakhulu e-AII-Mistral Large 4 iyimodeli ephambili eyethulwe yi-Mistral AI yase-France, enesiteketiso esithi “Le Chonk ikati elikhuluphele”. I-Mistral Large 4 isebenzisa ukwakheka kwe-MoE okunochwepheshe abahlukaniswe kahle, inamapharamitha angu-1.05T isiyonke, kuthi kusebenze cishe angu-49B kuphela ngesikhathi ngasinye, ine-encoder yombono engu-1.6B, inemultimodal yomdabu futhi inomongo ongu-1M. ...00
APUS-OpenJev-v1Amamodeli amakhulu e-AII-APUS-OpenJev-v1 iwuchungechunge lwamamodeli amakhulu okwenza izinqumo eceleni kwedivayisi olukhishwe njenge-open source yi-APUS AI Lab, olufana ne-Jev evaliwe ye-TypeSafe AI. Itholakala ngezisindo ze-4B, 9B kanye ne-35B-A3B, futhi ingasetshenziswa ekuzenzakaleleni kwesiphequluli, ...00
SemIfAmamodeli amakhulu e-AISemIf (eyayibizwa ngokuthi OpenJev) iyimodeli evulekile ye-AI yokwenza izinqumo, esebenzisa amamodeli avulekile ukuphindaphinda iphethini yesixhumi esibonakalayo sezinqumo zencazelo yesevisi evaliwe i-Jev ye-TypeSafe. I-SemIf ifunda ngokuqondile amaphuzu e-logits ahambisana nezinketho ezingamakhandidethi ekudluleni okukodwa okuya phambili...00
SocialCoachAmamodeli amakhulu e-AII-SocialCoach iyithuluzi lomqeqeshi we-AI lokuthuthukisa ukuhlakanipha kwemizwa, elikhishwe njengomthombo ovulekile ngokubambisana yi-Hong Kong University of Science and Technology (Guangzhou) kanye ne-Microsoft Research Asia. I-SocialCoach ivumela abasebenzisi ukuthi bazilolonge ngokuphelele izingxoxo okunzima ukuziqala nabalingiswa be-AI abanentukuthelo endaweni yokulingisa enobungozi obuphansi...00
KevUkuhlolwa kwamamodeli e-AIYini i-Kev? I-Kev iwumndeni wamamodeli okwenza izinqumo okhishwe njengomthombo ovulekile ngu-Jared Palmer, i-VP yezobunjiniyela kwa-Cognition, osekelwe ku-Qwen3.5/Qwen3.8 futhi onamamodeli amaningana kusukela ku-0.8B kuya ku-27B. I-Kev ayikhiqizi umbhalo; ifaka umbhalo wesimo kanye nemibuzo ehlelekile...00
T3POUmphakathi we-AI yomthombo ovulekileI-T3PO iyimodeli yokuhumusha ngokulandelana ngesikhathi sangempela ekhishwe yi-NetEase Youdao njengomthombo ovulekile, igama layo eliphelele lingu-“simulTaneous Translation via pareTo Policy Optimization”. I-T3PO igxile ekuhumusheni ngesikhathi umuntu esakhuluma, ihlola ngokuguqukayo ukuthi umongo wanele yini; uma wanele iqala ukuhumusha ngokushesha, uma unganele ilinda kancane, ...00
ArcLoopInkundla yokuthuthukisaI-ArcLoop iyinkundla yokukhiqiza amadrama opopayi asekelwe ku-AI, eyasungulwa nguQin Lu, owayengumholi wesayensi yedatha kwa-ByteDance. Inkundla yakhelwe phezu kwe-IP World, ihlanganisa abalingiswa, izigcawu, isitayela sokubukeka nezimpahla zemidiya kuphrojekthi eyodwa, futhi iqhubeke...00
GPT-6 LunaInkundla yokuthuthukisaI-GPT‑6 Luna iyimodeli elula eyethulwe yi-OpenAI, futhi iyinguqulo ephuma ku-GPT‑6 Astra kanye ne-GPT‑6 Sol, egxile emisebenzini evamile eyenziwa ngezinga elikhulu. Intengo ye-API yemodeli yehliselwe ku-$0.1 kokokufaka kanye no-$0.5 kokokuphum...00
WebCraftBenchUkuhlolwa kwamamodeli e-AII-WebCraftBench iyisisekelo sokuhlola ukukhiqizwa kwamakhasi ewebhu nge-AI, esethulwe yi-Tencent Hunyuan ngokubambisana ne-Tsinghua University ne-Peking University. I-WebCraftBench iqala ngokufaka ngokuzenzakalelayo i-instrumentation ezinhlelweni ezikhiqiziwe, bese ivumela ama-agent ukuthi ahlole imisebenzi yewebhu ngokuyisebenzisa ngempela nge-Playwright,...00
Open-RAILUmphakathi we-AI yomthombo ovulekileI-Open-RAIL iyisisekelo sobunjiniyela esijwayelekile sobuhlakani obugxile emzimbeni, esivulwe yi-China Mobile emhlabeni wonke, futhi iyisistimu yezinzwa ejwayelekile yokuqala embonini exhumanisa amamodeli e-VLA/WAM nemizimba yamarobhothi.00
ZDTaichu5.0-9BAmamodeli amakhulu e-AII-ZDTaichu5.0-9B iyimodeli entsha evulekile, ejwayelekile futhi enemodi eziningi ye-Zidong Taichu, eqondiswe emhlabeni obonakalayo nasezimweni zobuhlakani obugxile emzimbeni. Kwamabhentshimakhi omhlaba ayisishiyagalolunye omhlaba, ithole indawo yokuqala emhlabeni ezivivinyweni eziyisi-8 phakathi kwamamodeli anenani elifanayo lamapharamitha; i-MindCube-tiny ithole u-7...00
IrisAmamodeli amakhulu e-AIIris iyi-ejenti yokusesha ekhishwe ngaphansi kwelayisensi evulekile yiqembu le-AllSpark lase-Xiaohongshu, enezinguqulo ezimbili: i-35B (Iris-mini) ne-397B (Iris-pro). U-Iris anganquma ngokuzimela ukuthi usesheni, afunde kanjani, nokuthi aqede nini ukuphendula kuwebhu yangempela...00
书生-S2Amamodeli amakhulu e-AI书生-S2 iyimodeli enkulu eyisisekelo ye-multimodal (amapharamitha angu-397B) ekhishwe njenge-open source yi-Shanghai Artificial Intelligence Laboratory, futhi ihloselwe ukuba imodeli enkulu eyiqonda kakhulu isayensi. Imodeli isebenzisa i-Memory Decoder ukwethula amamojuli enkumbulo ochwepheshe axhunywayo, futhi inganweba ulwazi lwezizinda ngaphandle kokuqeqesha kabusha imodeli eyinhloko...00
PixRestoreUkuhlolwa kwamamodeli e-AII-PixRestore iyimodeli ehlanganisiwe yokulungisa izithombe eyethulwe ngokubambisana yi-Hong Kong Polytechnic University kanye ne-OPPO Research Institute. Imodeli isebenzisa i-Diffusion Transformer esikhaleni samaphikseli ukwenza i-flow matching ngqo ezingeni lamaphikseli, igweme ukulahleka kwemininingwane...00
AuKUmphakathi we-AI yomthombo ovulekileI-AuK iyimodeli eyisisekelo yokukhiqiza nokuhlela inkulumo enamapharamitha ayizigidi eziyizinkulungwane ezingu-1.5, ekhishwe njenge-open source yi-Tencent Hunyuan. Isebenzisa ukwakheka kwe-flow-matching diffusion esikhundleni se-autoregressive. Imodeli ingenza imisebenzi engu-16 yenkulumo ngesixhumi esisodwa semiyalo yolimi lwemvelo, okuhlanganisa i-zero-shot/instruct...00
DeepSeek V4.1 FlashInkundla yokuthuthukisaI-DeepSeek V4 Flash iyimodeli yolimi enkulu engasindi eyethulwe yi-DeepSeek, futhi yethulwa emphakathini ukuze ihlolwe nge-API ngoJulayi 31, 2025. Imodeli igxile esivinini nasekunciphiseni izindleko, ijubane lokucabanga lifinyelela kumakhulu ama-token ngomzuzwana, kanti ngezikhathi zokungasebenzi kakhulu, okukhiphayo lapho i-cache ishaya ku...00
MiniCPM5-2BAmamodeli amakhulu e-AII-MiniCPM5-2B iyimodeli eyisisekelo yolimi esebenza emadivayisini, ekhishwe ngomthombo ovulekile ngokubambisana yi-Mianbi Intelligent, umphakathi we-OpenBMB kanye ne-Tsinghua University. Imodeli inamapharamitha acishe abe yizigidi eziyizinkulungwane ezingu-2.5, isekela umongo ongu-128K, futhi ithole isilinganiso esingu-53.9 kumabhentshimakhi angu-34, yadlula amamodeli amaningi ezinga le-4B futhi yafinyelela ku-SOTA yamamodeli omthombo ovulekile ezinga le-2B...00
YoLiveInkundla yokuthuthukisaI-YoLive iyinkundla ka-Yoroll yokusakaza bukhoma izindaba ezisebenzisanayo ze-AI ngesikhathi sangempela. Ababukeli bavota ukuze banqume ukuthi indaba izoqhubeka kanjani esigcawini esilandelayo, kanti i-AI ikhiqiza izithombe ngokushesha isebenzisa imodeli ye-H3 Superfast. Ngemizuzwana engu-4 ingakhiqiza ividiyo engu-768p enomsindo ethatha imizuzwana engu-10...00
I-Spark X2.5Amamodeli amakhulu e-AII-Spark X2.5 iwuchungechunge lwamamodeli amakhulu ajwayelekile aphambili akhiqizwe yi-iFlytek, oluhlanganisa imodeli eyisisekelo ye-MoE esebenza efwini (293B/30B) kanye namamodeli omthombo ovulekile asebenza emishinini, okuhlanganisa i-Spark X2.5-4B ne-Spark X2.5-1.7B. Amamodeli aqeqeshwe ngokuphelele kusetshenziswa amandla ekhompyutha akhiqizwe eChina...00
I-National Anti-Fraud AIAmamodeli amakhulu e-AII-National Anti-Fraud AI iyinkundla yokufundisa nokuvimbela ukukhwabanisa kwezokuxhumana ngocingo naku-inthanethi, eqondiswa uMnyango Wamaphoyisa Womphakathi, i-Bureau of Criminal Investigation, futhi yathuthukiswa ngokuzimela yi-Shanghai Public Security Bureau. Le nkundla ihlanganisa amamodeli amakhulu olimi, amamodeli ezindlela eziningi kanye nobuchwepheshe bama-agent ahlakaniphile, inikeze izimpendulo ezihlakaniphile ze-AI, izindaba zokulwa nokukhwabanisa...00
E-Commerce BenchAmamodeli amakhulu e-AII-E-Commerce Bench iyisisekelo sokuhlola ukusebenza kwe-e-commerce kwesikhathi eside, esikhishwe njengomthombo ovulekile ngu-Alibaba Tongyi Qianwen ngokubambisana ne-Taotian Group. I-E-Commerce Bench ivumela i-LLM Agent ukuthi iphathe isitolo esiku-inthanethi izinsuku ezingu-365 ngemali yokuqala engu-100,000 yuan emakethe elingisa umhlaba wangempela,...00
Inkundla Evulekile ye-Alipay AIInkundla yokuthuthukisaInkundla Evulekile ye-Alipay AI iphethwe yi-Alipay (Hangzhou) Digital Service Technology Co., Ltd., futhi inikeza abathengisi, abahlinzeki bezinsizakalo nabathuthukisi bezinhlelo izindlela zokuxhuma amakhono e-Alipay, imibhalo yokuthuthukisa, amathuluzi kanye nokufinyelela kukhonsoli. Inkundla isekela izimo zokuthuthukisa ezifana nezinhlelo ezincane, izinhlelo zezinkampani zangaphandle, amawebhusayithi nezinhlelo zokusebenza zeselula, i-IoT, imidlalo emincane nama-akhawunti ezinsizakalo zokuphila, futhi isiqalile ukuhlola okuqondisiwe kweNkundla Evulekile ye-AI.068.5
TinyWowIbhokisi lamathuluzi e-AII-TinyWow iyibhokisi lamathuluzi e-AI eliku-inthanethi elenzelwe abafundi, othisha, ochwepheshe nabadali bokuqukethwe. Ihlanganisa amathuluzi e-PDF, ezithombe, amavidiyo, amafayela kanye nokubhala ngosizo lwe-AI. Abasebenzisi bangakhetha umsebenzi abawudingayo ngokuqondile ekhasini lewebhu, benze imisebenzi evamile efana nokuhlanganisa nokuguqula ama-PDF, ukucindezela nokuhlela amavidiyo, ukususa ingemuva ezithombeni, kanye nokukhiqiza noma ukubhala kabusha umbhalo. Inkundla igxile ekusetshenzisweni komuntu siqu okulula kwansuku zonke, futhi ilungele abasebenzisi abangafuni ukufaka isofthiwe yobungcweti. Ulwazi olutholakalayo lubonisa ukuthi amathuluzi ayo angasetshenziswa mahhala futhi akudingi ukubhalisa, kodwa amafomethi okufaka nemiphumela yokucubungula kungahluka kuye ngethuluzi. Okuqukethwe okukhiqizwe yi-AI kufanele kuhlolwe futhi kulungiswe ngumsebenzisi.038.3
FuturepediaIbhokisi lamathuluzi e-AII-Futurepedia iyikhathalogi nenkundla yokuthola amathuluzi obuhlakani bokwenziwa, ehlanganisa amathuluzi emikhakheni efana nokubhala, izithombe, amavidiyo nebhizinisi. Abasebenzisi bangaqala ngokusesha ngamagama angukhiye noma ngezigaba, bese behlunga ngokwemodeli yentengo, imisebenzi nokuthi i-API iyasekelwa yini. Bangabuka nezincazelo zamathuluzi, izimo zokusetshenziswa, imininingwane yamanani kanye nokubuyekezwa kwabasebenzisi. Le nkundla iqondiswe kochwepheshe, abathatha izinqumo emabhizinisini, abathuthukisi, abadali bokuqukethwe kanye nabathandi bobuhlakani bokwenziwa. Ilungele ukuthola amathuluzi angaba yizinketho, ukuqhathanisa izixazululo nokulandela izinguquko embonini. Kufanele kuqashelwe ukuthi imininingwane yekhathalogi nokubuyekezwa kwabasebenzisi kusafanele kuqinisekiswe kuwebhusayithi esemthethweni yethuluzi nangokulizama ngokoqobo.038.9
Llama 4Ibhokisi lamathuluzi e-AII-Llama 4 iyimodeli enkulu ye-multimodal yomthombo ovulekile eyakhishwa yi-Meta ngo-April 2025, esebenzisa ukwakheka kwe-multimodal komdabu kanye nemodeli yochwepheshe abaxubile (MoE). Ingacubungula umbhalo, izithombe, amavidiyo nomsindo, futhi inikeza izinhlobo ezifana ne-Scout ne-Maverick, esekela okokufaka komongo okufika cishe kuma-token ayizigidi ezingu-10. Onjiniyela, abacwaningi, amaqembu ezobuchwepheshe ezinkampani nabadali abazimele bangafaka imibhalo, ikhodi noma okuqukethwe kwe-multimedia kule modeli ukuze kuhlaziywe, kuphendulwe imibuzo, kufingqwe futhi kuthuthukiswe izinhlelo; imodeli iphinde isekele ukuthunyelwa kwasendaweni noma okuyimfihlo. Ukusetshenziswa kwangempela kusancike ohlotsheni oluthile, ekucushweni kwehadiwe nasendaweni yokuthunyelwa; umongo omude kakhulu awusho ukuthi yonke imisebenzi izogcina izinga elifanayo lokunemba.028
PostureCorrector.AIIbhokisi lamathuluzi e-AII-PostureCorrector.AI iyithuluzi lokuthola indlela yokuhlala elisekelwe ekuboneni ngobuhlakani bokwenziwa, elisebenzisa ikhamera evamile yekhompyutha ukubona izinguquko endleleni umsebenzisi ahlala ngayo phambi kwekhompyutha. Ngemva kokuvula ukutholwa ngekhamera, ithuluzi lizothumela isikhumbuzi ngelebula esikrinini noma ngesaziso sedeskithophu lapho lithola ukugoba phambili, ukugoba komhlane nezinye izimo zokuhlala ezingalungile, futhi liqophe izitayela zokuma komzimba. Umkhiqizo uphinde unikeze ngezikhumbuzi zokuphumula nokuphuza amanzi ngezikhathi ezibekiwe, futhi ungasetshenziswa ehhovisi, ekufundeni, ocwaningweni nasemidlalweni yekhompyutha yesikhathi eside. Ukubala kwenziwa kudivayisi yasendaweni, futhi kusekelwa ukusebenza ngaphandle kwe-inthanethi kanye nokulungiswa kwezinga lamafreyimu. Ufaneleka kakhulu njengethuluzi lokuqopha imikhuba yansuku zonke nokuthumela izikhumbuzi; ukusebenza kokutholwa kungase kuthinteke indawo yekhamera, ukukhanya nokufihlwa komzimba.047.5
Theee AIIbhokisi lamathuluzi e-AII-Theee AI iyinkundla ehlanganisa amathuluzi e-AI, futhi iwebhusayithi yayo ithi inikeza ukufinyelela endaweni eyodwa kumathuluzi e-AI angaphezu kuka-50,000, okusiza abasebenzisi bonge isikhathi abasichitha befuna amathuluzi kumawebhusayithi e-AI amaningi. Inkundla ibonisa izixhumanisi zamathuluzi ezigabeni ezinjengamavidiyo, ucwaningo, ukubala, idatha yewebhu, izibalo, izithombe, izethulo nokufunda. Abasebenzisi bangaphequlula iwebhusayithi bese bekhetha amathuluzi abawadingayo.087.8
falAma-API emidiya ekhiqizayoI-fal iyinkundla yemidiya ekhiqizwayo yabathuthukisi, ehlinzeka ngokufinyelela nokusabalalisa amamodeli ezithombe, amavidiyo, umsindo, umculo, izwi, i-3D kanye nokusakaza bukhoma. Abasebenzisi bangafinyelela amamodeli angaphezu kuka-1,000 nge-API eyodwa, noma basebenzise ama-GPU angenaseva, amaqoqo azinikele nezinsiza zekhompyutha ukuze bakhe, baqeqeshe futhi bakhulise izinhlelo zabo ze-AI.048
OfoxAIInkundla yokuthuthukisaI-OfoxAI inikeza isevisi ye-API yamamodeli amaningi yabathuthukisi namaqembu, esekela ukusebenzisa i-API Key eyodwa ukufinyelela kumamodeli angaphezu kuka-100, futhi ehambisana namaphrothokholi e-OpenAI, Anthropic ne-Gemini. Isevisi ihlanganisa izimo zokusebenzisa umbhalo, izithombe, amavidiyo nokuthuthukiswa kwama-Agent, ikhokhiswa ngokusetshenziswa kwangempela ngaphandle kwemali yesikhulumi. Izindawo eziyinhloko zokufinyelela zifaka i-API, imibhalo yabathuthukisi nekhonsoli.038
Roobin.aiIzixazululo ze-AI zamabhizinisiI-Roobin.ai ihlinzeka ngamasevisi amaqembu e-generative AI, ama-workshop e-design sprint, ukwenziwa ngcono kwesabelomali noxhaso oluphethwe lwamabhizinisi adinga amakhono e-AI enziwe ngokwezifiso, e-multimodal futhi avikelekile. Umkhiqizo ungahlanganiswa ne-Azure AI, uhlanganise yonke inqubo kusukela ekufakweni nasekusebenzeni kuya ekuhlanganisweni nezinqubo zebhizinisi, futhi usekele ukudalwa kokuqukethwe, ukuthuthukiswa kwesofthiwe, izinsiza zabantu, uxhaso lwe-IT nokuhlaziywa kwempendulo yamakhasimende.037.8
MiMoIbhokisi lamathuluzi e-AII-MiMo iyinkundla ejwayelekile yobuhlakani bokwenziwa eyethulwe yi-Xiaomi, ehlinzeka ngamakhono amamodeli agxile kubuchwepheshe bolimi, ezindlela eziningi kanye nezwi. Iwebhusayithi esemthethweni inikeza izindlela ezimbili eziyinhloko: i-Web Demo ne-API Access. Abasebenzisi bangaxhumana ngqo ne-MiMo, kuyilapho abathuthukisi bengahlanganisa amakhono ahlobene ngengosi yabathuthukisi ukuze bahlole ukubambisana phakathi kwabantu nemishini kanye nokwakhiwa kwezinhlelo zokusebenza ezihlakaniphile.078.5
FLUX 3Ibhokisi lamathuluzi e-AII-FLUX 3 iyimodeli eyisisekelo ye-multimodal eyethulwe yi-Black Forest Labs, futhi okwamanje isesigabeni se-Early Access. Imodeli ifunda izithombe, amavidiyo nomsindo ngokuhlanganyela ngaphakathi kwesakhiwo esihlanganisiwe, ukuze isetshenziswe ekudalweni kokuqukethwe, ekuqondeni okubukwayo, ekukhiqizeni amavidiyo nasekubikezeleni ukunyakaza. Ngokohlelo olusemthethweni, le makhono azotholakala kancane kancane nge-API, izisindo eziyimfihlo nezisindo ezivulekile. Okwamanje ungafaka isicelo sokufinyelela kwangaphambi kwesikhathi ekhasini elisemthethweni.008.3
Hugging FaceIbhokisi lamathuluzi e-AII-Hugging Face iyinkundla yokufunda komshini eyenzelwe abathuthukisi be-AI, ososayensi bedatha, abacwaningi nabafundi. Izinsiza zayo eziyinhloko zihlanganisa inqolobane yamamodeli, isikhungo samasethi edatha, umtapo we-Transformers kanye nama-Spaces. Abasebenzisi bangasesha imodeli noma isethi yedatha endaweni kuqala, bese beyibiza, benze izivivinyo futhi bayilungise kahle besebenzisa isixhumi esibonakalayo esifanayo; bangasebenzisa nama-Spaces ukubonisa izinhlelo ezisebenzisanayo. Uma kudingeka ukusatshalaliswa kokukhiqiza, bangafunda nangezinsizakalo zokuphetha ezisingathiwe. Le nkundla ilungele ukukhetha amamodeli, ukuqinisekisa ama-prototype, ukufundisa nocwaningo kanye nokuthuthukisa izinhlelo. Kufanele kuqashelwe ukuthi okuqukethwe kwenkundla kuvela kubanikeli abahlukene, ngakho ububanzi bokusetshenziswa, indawo encike kuyo kanye nemibandela yokusetshenziswa kwemodeli kungahluka; ngaphambi kokusetshenziswa kwangempela, kufanele kufundwe incazelo ehambisanayo.038.6
EaseMate AIIbhokisi lamathuluzi e-AII-EaseMate AI iyibhokisi lamathuluzi e-AI elenzelwe ukufunda, ukusebenza ehhovisi nokudala okuqukethwe. Ihlinzeka ngengxoxo enamamodeli amaningi, ingxoxo nemibhalo ye-PDF, ukubhala nokulungisa umbhalo nge-AI, ukuhumusha ngezilimi eziningi, ukusesha ku-inthanethi kanye nosizo lokudala olubonakalayo. Abasebenzisi bangakhetha imodeli noma ithuluzi elenzelwe isimo esithile kuwebhu noma esandisweni sesiphequluli, ukuze bakhiphe ulwazi emibhalweni, bakhe okuqukethwe, bahumushe umbhalo futhi bathole ulwazi. Ilungele abafundi, abacwaningi, abadali bokuqukethwe, ochwepheshe bezamabhizinisi kanye nabasebenzisi abadinga ukuxhumana ngezilimi ezahlukene. Umkhiqizo uhlinzeka ngezinketho zamahhala nezikhokhelwayo, futhi amamodeli nezici ezitholakalayo zingahluka kuye ngohlelo; okuqukethwe okukhiqiziwe, imiphumela yokuhumusha kanye nolwazi lokusesha kusadingeka kuqhathaniswe nemithombo yoqobo.037.6
StreamLakeIbhokisi lamathuluzi e-AII-StreamLake iwuphawu lwezinsizakalo zamabhizinisi olungaphansi kwe-Kuaishou, oluqondiswe kubathuthukisi bamabhizinisi namaqembu ebhizinisi adinga ubuchwepheshe bomsindo, ividiyo nobuhlakani bokwenziwa. Ihlinzeka ngezinsizakalo zamamodeli amakhulu, usizo lokuhlela nge-AI, izinsizakalo zamafu zomsindo nevidiyo, kanye nezixazululo ze-video AI. Ekusetshenzisweni, ngokuvamile kudingeka kukhethwe amandla afanele ngokuya ngesimo sebhizinisi, bese kusetshenziswa amathuluzi enkundla noma izixhumi zokuxhumana ukuze kuqedelwe ukuhlanganiswa. Lokhu kuhlanganisa ukusakaza bukhoma, ividiyo efunwayo, ukukhiqizwa kokuqukethwe, ukuqonda ividiyo nokuthuthukiswa kwezinhlelo ezihlakaniphile. Ilungele ezentengiselwano ze-inthanethi, amapulatifomu ezokuzijabulisa anhlobonhlobo, izikhungo zokusakaza nezemidiya ezintsha, kanye namabhizinisi endabuko aqhubezela phambili ukusetshenziswa kwe-AI. Kumele kuqashelwe ukuthi indlela yokuhlanganisa kanye nobubanzi bokusetshenziswa kwamandla athile kuncike kumojuli yomkhiqizo; amabhizinisi kusadingeka ahlole lokhu ngokuhambisana nesakhiwo sawo sobuchwepheshe nezidingo zebhizinisi.047.9
SaaS AI ToolsIbhokisi lamathuluzi e-AII-SaaS AI Tools iyinqolobane yezinsiza egxile ekuhlanganeni kwe-SaaS ne-AI, eyakhiwe futhi iphethwe ngokuzimela umsunguli uBren Kinfa. Le nkundla isebenzisa uhla oluhlukaniswe ngezigaba njengendawo yokuqala, iqoqe amathuluzi e-AI ekhiqizayo avela emhlabeni wonke, futhi inikeze izindaba zemboni ye-AI zansuku zonke, umphakathi wosomabhizinisi kanye nendawo yokuthumela amathuluzi. Lokhu kusiza abasebenzisi kusukela ekutholeni amathuluzi nasekuqondeni izinguquko zemboni kuya ekutholeni izinsiza zemikhiqizo. Ifanele abahloli bobuchwepheshe be-AI, osomabhizinisi, abaphathi bemikhiqizo, ochwepheshe bezokumaketha nabadali bokuqukethwe, kanye nabaklami nabathuthukisi. Le nkundla ngokwayo igxile kakhulu ekuhleleni ulwazi nasekutholeni izinsiza, hhayi ekunikezeni uhlelo oluthile lwe-AI; ukusebenza kwangempela, ukutholakala kanye nezindleko zamathuluzi afakiwe kusadinga ukuqinisekiswa ngumsebenzisi.018.2
ModelScopeIbhokisi lamathuluzi e-AII-ModelScope iyinkundla yezinsizakalo zamamodeli yomthombo ovulekile eyethulwe yi-DAMO Academy yakwa-Alibaba. Ihloselwe abathuthukisi be-AI, abacwaningi, amaqembu ezinkampani nezikhungo zemfundo, futhi inikeza uchungechunge lwamathuluzi olusuka ekuhloleni amamodeli luye ekwenzeni inference, ekuqeqesheni, ekusabalaliseni nasekuthuthukiseni izinhlelo zokusebenza. Le nkundla ihlanganisa amamodeli angaphezu kuka-300, ahlanganisa umbono wekhompyutha, ukucutshungulwa kolimi lwemvelo, inkulumo, ama-multimodal kanye ne-computing yesayensi, futhi inikeza namandla okwabelana ngamamodeli, ukuphatha ama-dataset kanye nokusabalalisa efwini. Abasebenzisi bangathola amamodeli ngokomsebenzi, bawahlole, bawenze i-fine-tuning noma bawaqeqeshe, bese bewaxhuma kumaphrojekthi besebenzisa umtapo wePython nezixhumanisi ezihlanganisiwe. Ekusetshenzisweni kwangempela, izindleko zokuhambisana kusafanele zihlolwe ngokusekelwe kokuncika kwemodeli, kumandla ekhompyutha nasendaweni yokusabalalisa.058.6
Toolify AIIbhokisi lamathuluzi e-AII-Toolify AI iyinkundla yokuqondisa yokuthola nokukhetha amathuluzi e-AI, ehlanganisa amathuluzi e-AI omhlaba wonke nesitolo sama-GPTs, futhi ihlele ikhathalogi ngokwezimo zokusetshenziswa, imisebenzi namathegi. Abasebenzisi bangathola amathuluzi ngokusesha ngamagama angukhiye, ukuphequlula izigaba noma ukubuka amazinga, babheke izethulo, inani lokuvakashelwa, izinga lokulondolozwa namaphuzu, bese bedlulela ekhasini elisemthethweni elihlobene. Ilungele abasebenzisi ngabanye, abadali bokuqukethwe, onjiniyela, kanye namaqembu namabhizinisi adinga ukwenza ucwaningo lwamathuluzi. Ikhathalogi nedatha ehlinzekwa yinkundla kungasetshenziselwa ukuhlunga kokuqala, kodwa ulwazi lwamathuluzi, amaphuzu nezixhumanisi zokufinyelela kungashintsha ngokuhamba kwesikhathi; ngaphambi kokuthenga ngokusemthethweni noma ukuhlanganisa ithuluzi ekuhambeni komsebenzi, kusafanele uvakashele ikhasi elisemthethweni ukuze uqinisekise.038.4
Malt AIIbhokisi lamathuluzi e-AII-Malt AI iyibhokisi lamathuluzi e-AI elenzelwe inqubo yokuthuthukiswa kwamaphrojekthi, elihlinzeka ngosizo oluqhubekayo ezidingweni, kuma-prototype, emibhalweni, ekwakhiweni kwedathabheyisi, ekubhalweni kwekhodi nasekuhlolweni okuzenzakalelayo. Abasebenzisi bangachaza imibono yebhizinisi ngolimi lwemvelo, bese bekhiqiza kancane kancane izinhlelo zama-prototype, imibhalo yephrojekthi kanye nesakhiwo sedathabheyisi, futhi bathole usizo ekubhalweni kwekhodi nasekuhlolweni. Ilungele abathuthukisi, abaphathi bemikhiqizo, amaqembu aqalayo, amaqembu ezinhlelo zangaphakathi ezinkampanini kanye namaqembu amaphrojekthi anemisebenzi ehlukahlukene, ngenhloso yokunciphisa ukushintshashintsha phakathi kokudluliswa kwezidingo nezigaba zokuthuthukisa. Kumaphrojekthi asuka emcabangweni aye ekusetshenzisweni, i-Malt AI ingasetshenziswa njengethuluzi lokubambisana, kodwa ama-prototype, izakhiwo, ikhodi nemiphumela yokuhlolwa ekhiqiziwe kusadingeka kuhlolwe, kulungiswe futhi kuqinisekiswe yiqembu ngokwezidingo zangempela.0128.3
Liquid AIIbhokisi lamathuluzi e-AII-Liquid AI iyibhokisi lamathuluzi e-AI elenzelwe amadivayisi asemaphethelweni, umkhiqizo wayo oyinhloko i-Liquid Foundation Models (LFMs), enosayizi abasuka kumapharamitha angu-350M kuye ku-2.6B futhi esekela izinhlobo zamamodeli ombhalo, olimi nombono kanye nomsindo. Abathuthukisi bangasebenzisa inkundla ye-LEAP ukulanda amamodeli, ukuwenza kahle nokuwasebenzisa kumadivayisi, noma basebenzise uhlelo lweselula i-Apollo ukuhlola endaweni. Le nkundla ihloselwe abathuthukisi abazimele, amabhizinisi asaqalayo, abacwaningi, abakhiqizi bezingxenyekazi zekhompuyutha kanye namabhizinisi adinga izixazululo ezenziwe ngokwezifiso, futhi ifanele amatheminali afana namafoni, amakhompyutha aphathekayo, izimoto namarobhothi. Amamodeli atholakala ku-Hugging Face; imiphumela ethile isancike ezimweni zehadiwe, kusayizi wemodeli nasekuhambisaneni kokusetshenziswa.057.8
LlamaFactory OnlineIbhokisi lamathuluzi e-AII-LlamaFactory Online iyibhokisi lamathuluzi e-AI eliku-inthanethi elenzelwe ukulungisa kahle amamodeli amakhulu avulekile, ikakhulukazi lisiza onjiniyela abazimele, amabhizinisi aqalayo namaqembu ocwaningo emanyuvesi. Abasebenzisi bangalayisha idatha ngesixhumi esibonakalayo sewebhu, bakhethe imodeli, indlela yokuqeqesha nohlelo lokunemba, bese benza ukuqeqesha, ukuqapha nemisebenzi ehlobene nokusebenzisa imodeli, ngaleyo ndlela behlise umsebenzi wokulungisa indawo ye-GPU yasendaweni. Inkundla isekela amamodeli amaningi amakhulu avulekile asetshenziswa kakhulu, ukuqeqeshwa kwangaphambilini, i-SFT, i-Reward Modeling, i-PPO, i-DPO, i-KTO nezinye izindlela zokuqeqesha, futhi inikeza izinketho zokulungisa kahle ze-LoRA, i-QLoRA nokulungisa yonke imingcele. Inkundla inezindlela zamahhala nezikhokhelwayo; ukusetshenziswa kwangempela kusadinga ukuhlola ukusetshenziswa kwamandla ekhompyutha nezindleko zokufunda kuye ngosayizi wedatha, uhlobo lwemodeli nokulungiselelwa kokuqeqesha.058
deepkeyAma-API amamodeli amakhuluI-deepkey iyisango le-API lamamodeli amakhulu elenzelwe ukuhlela nge-AI, ukuhlola amamodeli nokuthuthukisa izinhlelo zokusebenza. Nge-API eyodwa, ixhuma izinsiza zamamodeli ezifana ne-Claude, OpenAI ne-Gemini. Ngemva kokubhalisa, abasebenzisi bangakha ukhiye we-API, bashintshe i-BASE_URL ngokwemibhalo, bese bexhuma amamodeli kumathuluzi afana ne-Claude Code, Codex CLI, Gemini CLI, VSCode noma Cursor. Inkundla ibonisa ndawonye imininingwane yamamodeli, abahlinzeki, izindlela zokukhokhisa nama-endpoint, futhi inikeza imibhalo yokufaka ku-Windows, ukulungisa iziguquguquko zemvelo, ukuthuthukisa, ukusula inqolobane nokuxazulula izinkinga zokuxhuma. Ilungele abathuthukisi, abahloli namaqembu ezobuchwepheshe adinga ukushintsha amamodeli phakathi kwamathuluzi amaningi. Ngaphambi kokuyisebenzisa, qinisekisa ukuthi ithuluzi olikhonzile lisekela ama-endpoint enziwe ngokwezifiso, bese uhlelela izindleko zokusebenzisa ngokuvumelana namanani enkundla.067.5
Moodboard CreatorIbhokisi lamathuluzi e-AII-Moodboard Creator iyithuluzi le-AI lokukhiqiza ama-moodboard elenzelwe abaklami bezimpawu zemikhiqizo, be-UI/UX, bangaphakathi nezemfashini. Ngemva kokufaka amagama angukhiye noma incazelo yesitayela, uhlelo lungakhiqiza i-moodboard enezinto ezifana nemibala, amafonti nezithombe cishe emizuzwini emi-3, futhi luvumele abasebenzisi ukuthi baqhubeke nokulungisa izinto nokuhlelwa kwazo. Ithuluzi liphinde linikeze ugqozi lwama-logo neziphakamiso zokuhlanganisa amafonti, lisekele ukufanisa izithombe, ukwakheka nezimpawu ezivela kulabhulali yezinto, futhi likwazi ukuthekelisa ku-Figma, Canva nakwezinye izinkundla zokuklama ukuze kusetshenziswane yiqembu futhi kuqoqwe impendulo. Lifanele ukuhlola isitayela nokulungiselela iziphakamiso ekuqaleni kwephrojekthi, kodwa okuqukethwe okukhiqiziwe kusadinga ukuhlungwa nokulungiswa ngokwezidingo ezithile zohlobo lomkhiqizo.017.9
AI Tool HuntIbhokisi lamathuluzi e-AII-AI Tool Hunt iyikhithi yamathuluzi e-AI kanye nenkundla yokuwathola, equkethe izinhlelo zokusebenza ezingaphezu kuka-18,000 ezigabeni ezingaphezu kuka-120, ezihlanganisa ukubhala, izithombe, amavidiyo, ukuhlela izinhlelo, ezamabhizinisi nezinye. Abasebenzisi bangasesha amathuluzi besebenzisa amagama angukhiye, noma bawahluze ngokwenani, isigaba, amaphuzu nezinye izimo, bese besebenzisa ukubuyekezwa kwabasebenzisi ukuqhathanisa imikhiqizo ehlukene. Inkundla iphinde inikeze umsebenzi wokulondoloza, ukuze kube lula ukuhlela amathuluzi athakazelisayo; abathuthukisi bangathumela imikhiqizo yabo ye-AI. Ilungele abadali bokuqukethwe, abaklami, abathuthukisi, abasebenzi bezokumaketha, abanikazi bamabhizinisi, abafundi nabacwaningi abafuna ukuthola amathuluzi. Qaphela ukuthi ulwazi lwemibhalo, ukubuyekezwa kwabasebenzisi, kanye nemisebenzi namanani amathuluzi ngokwawo kungashintsha; ngaphambi kokuwasebenzisa, kusafanele ubheke ikhasi lomkhiqizo elihambisanayo.018.2
ToolfolioIbhokisi lamathuluzi e-AII-Toolfolio iyibhokisi lamathuluzi e-AI lamahhala kanye nohlu lwezinsiza oluqhutshwa ngokuzimele, oluqondiswe kubaklami, abathuthukisi, abadali bokuqukethwe, abaphathi bemikhiqizo, abasebenzi bezokumaketha kanye nabathuthukisi abazimele. Inkundla ihlela amathuluzi ngokwemikhakha efana nokuklama, ukuthuthukisa nezinhlelo ze-AI, futhi ihlanganisa nama-UI Kits, izifanekiso nama-plugin ahlobene ne-Figma ne-Webflow, kanye nezinsiza zokuklama ezifana namaphalethi emibala, amaqoqo amafonti namaqoqo ezithonjana. Abasebenzisi bangaqala ngokuphequlula ngokwezigaba, bese bebuka futhi behlunga izinsiza ngokwezidingo zephrojekthi; inkundla iphinde ishicilele izibuyekezo zamathuluzi, izinsiza zamahhala nolwazi oluhlobene ngezincwadi zezindaba zamasonto onke nangebhulogi. Lolu hlu lufanele ucwaningo lokuqala, ukungeza ekusebenzeni kanye nokuqoqwa kwemibono, kodwa okuqukethwe okufakiwe akusho ukuthi kuyisincomo esiqinisekile; ngaphambi kokulusebenzisa, abasebenzisi kusafanele baqinisekise ngokwabo imisebenzi yethuluzi, ukuhambisana kwalo nesimo salo samanje sokutholakala.038.2
AI ArenaIbhokisi lamathuluzi e-AII-AI Arena iyinkundla evulekile yokulinganisa abasizi bokuhlela be-AI nama-ejenti okubhala ikhodi, eyasungulwa yi-JetBrains futhi ehlelelwe ukunikelwa ekuphathweni yi-Linux Foundation. Inkundla yakha inqubo yokuhlola egxile emisebenzini yangempela yokuthuthukisa isofthiwe, ehlanganisa ukulungiswa kwama-patch, ukulungiswa kwamaphutha, ukubuyekezwa kwekhodi nokukhiqizwa kwezivivinyo. Abahlinzeki bamathuluzi e-AI, abahlinzeki bezobuchwepheshe, abasebenzisi bamabhizinisi kanye nomphakathi wabathuthukisi bangasebenzisa amapayipi okuhlola, amasethi edatha nemithetho evulekile ukuqhathanisa amathuluzi ahlukahlukene, futhi bangafaka isandla ngedatha yokuhlola nezinhlelo zokulinganisa. Inkundla isekela ukusebenzisa amasethi edatha akho ukuze kuqhutshwe ukuhlola ezikhungweni zasendaweni. Iziphetho zokuhlola zithonywa yisethi yedatha, izilungiselelo zemisebenzi nemithetho yokukala, ngakho azikwazi ukuthatha indawo yokuqinisekiswa kwangempela kweqembu elithile.017.1
Inkundla Evulekile ye-Baidu AIIbhokisi lamathuluzi e-AIInkundla Evulekile ye-Baidu AI iqondiswe kubathuthukisi, amabhizinisi, izikhungo zocwaningo nabahlanganisi bezimboni, futhi ihlinzeka ngamakhono e-AI afana nenkulumo, umbono wekhompyutha, ukucutshungulwa kolimi lwemvelo namagrafu olwazi, kanye nezinsizakalo zamamodeli amakhulu agxile ochungechungeni lwe-Wenxin. Abasebenzisi bangabiza ama-API noma bahlanganise ama-SDK ngokwezidingo zebhizinisi, baxhumanise amakhono afana nezimpendulo ezihlakaniphile, ukudalwa kokuqukethwe, ukusebenzisana ngenkulumo nokubonwa kwezithombe ezinhlelweni zabo, noma basebenzise ukwesekwa kwenkundla ekuthuthukiseni, ekuqeqesheni amamodeli nasekuwasebenziseni. Inkundla ihlanganisa nezimo ezifana nezingxenyekazi zekhompuyutha ezihlakaniphile, ezempilo nezokwelapha, kanye nezinsizakalo zamabhizinisi. Ekusetshenzisweni kwangempela, ukuthuthukiswa kufanele kuhambisane nemibhalo yama-API, ukufaneleka kwemodeli nezidingo zedatha yebhizinisi; amaphrojekthi ayinkimbinkimbi asengadinga ukuhlanganiswa kobunjiniyela nokuhlolwa kwemiphumela okwenziwa yithimba ngokwalo.028.4
AI帮个忙Ibhokisi lamathuluzi e-AII-AI帮个忙 iyinkundla enemisebenzi eminingi yokwandisa ukusebenza kahle, eqondiswe kubasebenzisi abakhuluma isiShayina. Ihlanganisa amathuluzi ezimo ezifana nokudala okuqukethwe, umsebenzi wasehhovisi, usizo lokufunda, ezokuzijabulisa nokuphila kwansuku zonke. Abasebenzisi bangakhetha ithuluzi elifanele ngokuya ngomsebenzi, bafake isihloko, izidingo noma izinto eziyisisekelo, bese bekhiqiza imibhalo, izinhlelo, izimpendulo nemiphumela yokuhlaziya, abayilungise ngokuvumelana nokusetshenziswa kwangempela. Le nkundla ilungele abadali bokuqukethwe, abasebenzi basehhovisi, abafundi kanye nabasebenzisi abanezidingo ezincane zokwandisa ukusebenza kahle kwansuku zonke. Ihlanganisa izimo zokuqukethwe kwesiShayina ezifana ne-Xiaohongshu ne-Weibo, kanye nemisebenzi efana nezifinyezo zemibiko yamasonto onke, ukuthuthukisa ama-CV, ukuhlela ulimi lwamaphepha ocwaningo, ukuhumusha nokulungisa ikhodi. Okuqukethwe okukhiqiziwe kusadinga ukuqinisekiswa ngumuntu, kanti eminye imisebenzi efana nokunciphisa ukufana kwezihloko, ukunweba okuqukethwe koqobo kanye nemibiko ejulile ingamalungelo amalungu.048.2
TopaiIbhokisi lamathuluzi e-AII-Topai iyikhathalogi yamathuluzi egxile ekutholeni nasekuhlungeni izinhlelo zokusebenza ze-AI, ihlela izinkulungwane zamathuluzi ngezigaba ezifana nezithombe, umsindo, ukubhala nokuthuthukisa izinhlelo. Abasebenzisi bangathola amathuluzi ngokuphequlula izigaba noma ngokusesha, babuke izibuyekezo ezifanele, balondoloze amaphrojekthi abanentshisekelo kuwo, futhi baphequlule amathuluzi amasha, izinhlu ezihlelwe ngezindikimba nemihlahlandlela yama-prompt. Inkundla ilungele abaklami, abathuthukisi, ochwepheshe bezokumaketha, abadali bokuqukethwe, abaphathi bamabhizinisi, abafundi, abacwaningi nabathandi be-AI abafuna amathuluzi asizayo ekudaleni, ekufundeni nasekuhleleni izinqubo zebhizinisi. Ikhathalogi ngokwayo ihlinzeka kakhulu ngenkomba yolwazi; ukusebenza, izindleko, ukuzinza kwesevisi nendlela yokuphatha idatha yamathuluzi athile kusafanele kuqinisekiswe ngumsebenzisi ngokuzimela ngaphambi kokuwasebenzisa.018.2
AI LibraryIbhokisi lamathuluzi e-AII-AI Library ingumtapo wolwazi wamahhala wokuthola amathuluzi obuhlakani bokwenziwa, ohlanganisa amathuluzi nezinsiza zamanethiwekhi ezinzwa emikhakheni efana nombhalo, izithombe, i-3D, umsindo nevidiyo. Abasebenzisi bangaqala ngokuhlunga ngokomsebenzi, ngentengo, ngohlobo lokufaka nokukhipha noma ngomkhakha, bese bebuka izinhlelo zokusebenza zeWeb, amaphrojekthi e-Google Colab nawe-HuggingFace ahambisanayo, futhi bahlele izinsiza abazisebenzisa njalo ngamabhukhimakhi namanothi abo. Umtapo wolwazi uphinde unikeze ngezinto zokufunda ezingaphezu kuka-200, ezifanele abadali bokuqukethwe, abakhangisi, onjiniyela, ochwepheshe abasebenzayo nabafundi ekucwaningeni amathuluzi, ekusizeni ekudaleni nasekufundeni. Qaphela ukuthi i-AI Library ngokuyinhloko isebenza ekutholeni nasekuhleleni izinsiza; izimo zokufinyelela, indlela yokusebenzisa nemiphumela yamathuluzi athile kusamele kuqinisekiswe endaweni yokuqala.018.1
ToolScoutIbhokisi lamathuluzi e-AII-ToolScout iyibhokisi lamathuluzi e-AI kanye nenkundla yokuhlanganisa amathuluzi, eyenzelwe abasebenzisi abafuna ukuthola, ukuqhathanisa nokuqonda izinhlelo zokusebenza ze-AI. Le nkundla ihlanganisa amathuluzi emikhakheni efana nokubhala, ukuklama nokuhlela izinhlelo, futhi inikeza izinketho zokusesha nokuhlunga ngokwezigaba, amapulatifomu, amanani nezinye izimo. Abasebenzisi bangabheka izincazelo zamathuluzi, izibonelo zokusetshenziswa kanye nokubuyekezwa kwabasebenzisi, futhi bafunde izindaba ze-AI zansuku zonke ukuze bathole umbono wokuqala wokuthi ithuluzi lingasetshenziswa kuphi. Ilungele abathandi bobuchwepheshe be-AI, abadali bokuqukethwe, ochwepheshe, abafundi nabacwaningi. Qaphela ukuthi ulwazi lwemibhalo, ukubuyekezwa kanye nesimo sesevisi samathuluzi kungashintsha; ngaphambi kokusebenzisa ithuluzi ngokusemthethweni, kusafanele ubheke imininingwane yemisebenzi namanani ekhasini lethuluzi elithile.028.5
AI ValleyIbhokisi lamathuluzi e-AII-AI Valley iyibhokisi lamathuluzi elenzelwe ukutholwa nokuseshwa kwamathuluzi e-AI, elihlanganisa izinhlelo zokusebenza ze-AI namaprompthi avela emhlabeni wonke, ahlelwe ngezigaba ezifana nokuklama, ukuhlela amakhodi nabasizi bempilo yansuku zonke. Abasebenzisi bangaphequlula izigaba kuqala noma bafake izidingo zabo ekusesheni, bese bekhetha amathuluzi abangawazama ngokusekelwe ezincazelweni zawo. Inkundla uqobo imahhala, futhi ilungele abaqalayo ku-AI, abathuthukisi, abaklami, abadali bokuqukethwe, abafundi nabacwaningi abafuna amathuluzi okusiza ekubhaleni, ekuklameni, ekuhleleni amakhodi nasekuhleleni ulwazi. Qaphela ukuthi amathuluzi akuhlu ahlinzekwa abahlinzeki bezinsizakalo abahlukene; izici ezithile, imibandela yokusetshenziswa kanye nokuthi ayakhokhelwa yini kufanele kuqinisekiswe ekhasini lethuluzi elifanele.018.2
Arena AIIbhokisi lamathuluzi e-AII-Arena AI iyinkundla eku-inthanethi yokuncintisana nokuxoxa ngamamodeli amakhulu obuhlakani bokwenziwa, eyakhiwe ithimba labacwaningi laseNyuvesi yaseCalifornia, eBerkeley. Abasebenzisi bangathumela umbuzo, benze amamodeli amabili angaziwa aphendule ngesikhathi esisodwa bese bevota, noma bakhethe amamodeli amabili ukuze bawaqhathanise ngapha nangapha, noma bazizwele imodeli eyodwa. Inkundla ibuyekeza amazinga amamodeli ngokusekelwe emiphumeleni yamavoti ezingxoxo, futhi ifanele abacwaningi bama-algorithm, onjiniyela bezaziso, abadali bokuqukethwe kwezobuchwepheshe kanye nabasebenzisi abafuna ukuqhathanisa ukusebenza kwamamodeli. Inkundla inikeza ukufinyelela kwamahhala, kodwa amazinga athonywa uhlobo lwemibuzo, ulimi, izintandokazi zabahlanganyeli kanye nobukhulu besampula, ngakho ayinakuthatha indawo yokuhlolwa kochwepheshe okuqondene nomsebenzi othile.097.6
LMArenaIbhokisi lamathuluzi e-AII-LMArena iyinkundla yokuhlola amamodeli obuhlakani bokwenziwa eqhutshwa umphakathi, eholwa ithimba lezemfundo lase-University of California, Berkeley. Inqubo yayo eyinhloko iwukuthi amamodeli amabili anamagama afihliwe aphendule umbuzo ofanayo, bese abasebenzisi bevota ngokusekelwe kwikhwalithi yezimpendulo; ubunikazi bamamodeli buboniswa ngemva kokuvota. Inkundla isebenzisa imiphumela yamavoti kanye nohlelo lwamaphuzu i-Elo olufana nolusetshenziswa ku-chess ukuze yakhe izikhundla eziguqukayo, futhi ihlanganisa imisebenzi yombhalo, izithombe neminye. Ilungele abacwaningi bobuhlakani bokwenziwa, onjiniyela, abathanda ubuchwepheshe kanye nabasebenzisi abajwayelekile abafuna ukuqhathanisa amamodeli ngokuvundlile. Iphinde inikeze ngamasethi edatha avulekile azosetshenziswa ocwaningweni. Kumele kuqashelwe ukuthi izikhundla zibonisa okuthandwa ngabasebenzisi bangempela kanye nokusebenza emisebenzini ethile, futhi azikwazi ukuthatha indawo yokuhlolwa okukhethekile okuqondene nezidingo zebhizinisi, izindleko noma ukuzinza.048
聚好用 AIIbhokisi lamathuluzi e-AI聚好用 AI yibhokisi lamathuluzi e-AI elihlanganisiwe elenzelwe ukudalwa kokuqukethwe, elihlanganisa imisebenzi yezithombe, umculo, amavidiyo nokubhala. Abasebenzisi bangasebenzisa isixhumi esibonakalayo sesiShayina ukukhetha ithuluzi elifanele ngokomsebenzi, bese beqedela izinqubo zokukhiqiza izithombe, ukwenza amavidiyo, ukucubungula umbhalo noma ukudala umsindo. Le nkundla ilungele abasebenzi bokudala, amaqembu okuqukethwe ezinkampani kanye nabaqalayo abafuna ukuzama ukudala besebenzisa ubuhlakani bokwenziwa. Ingase futhi isetshenziselwe ukukhiqiza okusalungiswa kokuqala kwezinto zokumaketha, imiqondo ebonakalayo kanye nemibhalo yansuku zonke. Le nkundla inikeza izindlela zamahhala nezikhokhelwayo; ububanzi bemisebenzi, imikhawulo yokusetshenziswa kanye nemithetho yezindleko kufanele kuqinisekiswe ekhasini langempela.017.6
DeepAIIbhokisi lamathuluzi e-AII-DeepAI iyibhokisi lamathuluzi e-AI elihlanganisa imisebenzi eminingi yokukhiqiza, elisebenzela kakhulu abadali bokuqukethwe, abaculi, abaklami, ochwepheshe bezokumaketha kanye nabathuthukisi abadinga ukusebenzisa amandla e-AI. Abasebenzisi bangathola usizo lokudala imibhalo ngengxoxo, basebenzise izincazelo zombhalo ukwenza izithombe noma amavidiyo, futhi bahlele izithombe endaweni eyodwa; abathuthukisi bangasebenzisa i-API ukufaka la mandla ezinhlelweni zabo. Inkundla inikeza inguqulo yamahhala kanye nokubhalisa okukhokhelwayo, kanti i-API isekela ukukhokhiswa ngokusetshenziswa. Ilungele ukwakha imibono ngokushesha nokwenza izinto zokuqala, kodwa imiphumela ekhiqiziwe isadinga ukuhlolwa nokulungiswa ngumuntu, kanye nokuqinisekiswa kokuthi iyahambisana yini nezidingo zokusetshenziswa okuthile.028.5
MagicArenaIbhokisi lamathuluzi e-AIIMagicArena iyibhokisi lamathuluzi e-AI eliqondiswe kubadali bokuqukethwe okubukwayo, onjiniyela bezikhuthazo, abacwaningi namaqembu athuthukisa okuqukethwe. Inkundla ibeka amamodeli amaningi amakhulu okubona enqubweni efanayo yokuqhudelana. Ngemva kokufaka isikhuthazo esifanayo, abasebenzisi bangaqhathanisa izithombe noma amavidiyo akhiqizwe amamodeli ahlukene futhi babambe iqhaza ekuhloleni ngokuvota okungaziwa. Inkundla isekela ukubuka amarekhodi omlando emiqhudelwano agcinwe efwini, okwenza kube lula ukubuyekeza imiphumela yokukhiqiza nokuhlaziya umehluko phakathi kwamamodeli. Ilungele ukuhlunga kokuqala amamodeli, ukuhlola izikhuthazo nokuqhathanisa imiphumela ngaphambi kokudala. Inkundla inikeza izindlela zamahhala nezikhokhelwayo, futhi ububanzi obutholakalayo bungahluka; imiphumela yomqhudelwano nayo ithonywa yizikhuthazo, uhlobo lomsebenzi, inguqulo yemodeli kanye nesampula yamavoti, ngakho ayikwazi ukuthatha indawo yokuhlolwa okusemthethweni kwephrojekthi.028.1
I-AI Navigation ka-YouSheIbhokisi lamathuluzi e-AII-AI Navigation ka-YouShe iyibhokisi lamathuluzi e-AI eliqondiswe kubaklami, abasebenzi bokudala kanye nabathandi be-AI, elisetshenziswa kakhulu ukuthola amathuluzi afanele ngokwezigaba nezinkambu zobungcweti. Le nkundla ihlanganisa amathuluzi angaphezu kwezigaba eziyishumi, okuhlanganisa ukudweba nge-AI, ingxoxo nge-AI, ukubhala okuhlakaniphile nokucubungula umsindo nevidiyo, futhi inweba iye ezinkambu ezikhethekile ezifana nokuklama i-UI, ukuklama kwezithombe nokuklama kwe-3D. Abasebenzisi bangaqala ngokuphequlula izigaba ngokwezidingo zabo, bese besebenzisa izinsiza ezifana nezifundo zokuklama, izinto zokusetshenziswa, imitapo yezithombe enokucaca okuphezulu nezikimu zemibala ukuze baqhubeke nokudala. Bangase futhi basebenzise ukungena komphakathi ukuze baxhumane futhi bafunde. Le nkundla isebenza mahhala futhi ibuyekezwa ngokuqhubekayo. Qaphela ukuthi umsebenzi wayo oyinhloko ukuhlela ulwazi nokusiza ekutholeni amathuluzi; ukutholakala, imithetho yokusetshenziswa nezinguquko zesevisi zamathuluzi athile kusancike ezinkundleni ezihambisanayo.078.7
PoeIbhokisi lamathuluzi e-AII-Poe iyinkundla ehlanganisa amamodeli obuhlakani bokwenziwa eyakhiwe yi-Quora, egxile ekuzameni nasekuqhathaniseni amamodeli amaningi obuhlakani bokwenziwa kusevisi eyodwa. Abasebenzisi bangashintsha phakathi kwamamodeli ahlukene, benze izingxoxo noma bakhiqize okuqukethwe, futhi badale, balungiselele futhi babelane ngamarobhothi enzelwe izinhloso ezithile. Inkundla iyatholakala kuwebhu nakumadivayisi eselula, futhi inikeza amakhono ahlobene nombhalo, izithombe, amavidiyo nomsindo. Ilungele abathandi bezobuchwepheshe, abathuthukisi, ababhali, abakhangisi nabafundi. I-Poe inikeza izindlela zokusebenzisa zamahhala nezikhokhelwayo; amamodeli nezici ezitholakalayo kufanele kuqinisekiswe ekhasini lamanje lenkundla. Amamodeli ahlukene anekhwalithi yokukhiphayo nemisebenzi afaneleka kuyo engafani, ngakho imiphumela ekhiqiziwe kusafanele ihlolwe umuntu.018.5
AtlantixInkundla yokuthuthukisaI-Atlantix iyinkundla yokuthola, ukuthuthukisa nokwenza izinto ezintsha zibe yimikhiqizo noma izinsizakalo ezithengiswayo. Iguqula imiphumela emisha evela kososayensi abahamba phambili nasemanyuvesi ibe yizixazululo ezilungele imakethe. Abasebenzisi bangakhetha ubuchwepheshe kusizindalwazi esikhethiwe, basebenzise i-AI ukwenza uhlelo lwebhizinisi, futhi baxhumane nochwepheshe, abatshalizimali nabeluleki—konke kuhlanganiswe enkundleni eyodwa.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.00
SoniloInkundla yokuthuthukisaI-Sonilo iyinkundla ye-AI yokukhiqiza umculo ohambisana namavidiyo nemisindo, eyasungulwa ithimba eliyinhloko langaphambili le-ByteDance ne-Mureka yase-Kunlun Tech. Ngemva kokulayisha ividiyo, i-AI ihlaziya ngokuzenzakalela isigqi, imizwa nezenzo ezisesithombeni, bese ikhiqiza umculo nemisindo ehambisanayo...00
Eye2.aiUkuhamba kwezinto ze-AIBuza umbuzo owodwa, uqhathanise izimpendulo zama-AI adumile zibhekene, ubone lapho zivumelana khona.00
Tattoo CanvasUkuhamba kwezinto ze-AII-Tattoo Canvas inikeza iqoqo elikhethiwe logqozi lwama-tattoo, elihlanganisa izitayela ezihlukahlukene ezifana nemidwebo, ubuncane, ezendabuko, ezaseJapane nezibukeka njengezangempela. Phequlula amagalari anezindikimba ukuze uthole ugqozi, funda ngomklamo ngamunye njengereferensi, noma dala umqondo we-tattoo owenziwe ngokwezifiso ekhasini lokukhiqiza uma udinga umqondo omusha ngokuphelele. Imisebenzi ekhiqizwe ngemva kokungena ngemvume igcinwa kokuthi «Imiklamo Yami», ukuze ukwazi ukuyibuka futhi noma nini.00
PixelRAGInkundla yokuthuthukisaI-PixelRAG iwuhlaka lwe-RAG olugxile ekuboneni, olukhishwe njenge-open source yi-SkyLab/BAIR yase-Berkeley. Esikhundleni sendlela evamile yokukhipha amakhasi ewebhu abe umbhalo omsulwa bese kuwasesha, uhlaka lusebenzisa isiphequluli ukuguqula amakhasi ewebhu nama-PDF abe amathayela ezithombe-skrini, lusebenzisa i-Qwen3-VL ecushwe nge-LoRA...00
Ling-3.0-tiny – imodeli yokucabanga exubile yemvelo eyethulwe yi-Ant LingAmamodeli amakhulu e-AII-Ling-3.0-tiny – imodeli yokucabanga exubile yemvelo eyethulwe yi-Ant Ling – iyimodeli esebenzisa ukwakheka kochwepheshe abaxubile abangenazinsiza (MoE), enamapharamitha angu-7.9B esewonke futhi esebenzisa cishe angu-1.3B kuphela ithokheni ngayinye, ihlose ukulinganisa kahle kokubala namandla aqinile okuveza ulwazi. Imodeli isekela ukubiza amathuluzi ngokwemvelo, futhi ingaqondisa ngokuqondile ukulawulwa kwe-UI yesiphequluli kanye nemisebenzi yokuzenzakalela kumadivayisi eselula. Iphinde isekele ukucabanga okwenziwa ngokuphelele endaweni, ngakho ingasebenza ngaphandle kokuncika efwini. Izinzuzo zayo eziyinhloko zihlanganisa ukukhanya ngokwedlulele, ukubeka ubumfihlo kuqala kanye namandla emvelo okusebenzisa amathuluzi, okuyenza ifaneleke kakhulu izimo zokuthunyelwa kumadivayisi asemaphethelweni nasemaphethelweni enezinsiza ezilinganiselwe. Ukusetshenziswa okujwayelekile kufaka abasizi abahlakaniphile abangaxhunyiwe ku-inthanethi kumadivayisi eselula, ukuzenzakalela kweziphequluli, ukuphathwa kolwazi lwasendaweni (njengokuhlanganiswa ne-obsidian-cli), kanye nezinsizakalo ze-AI ezithobela imithetho ezinkampanini ezisebenzisa amanethiwekhi angaphakathi noma amakhompyutha okulawula izimboni.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.00
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.00
I-E-Bench – ibhentshimakhi yokuhlola ama-agent ehlakaniphile eyethulwe yi-Tencent Hunyuan nabanyeUkuhlolwa kwamamodeli e-AII-E-Bench iyibhentshimakhi yokuhlola ukusetshenziswa kwamathuluzi ngezinyathelo eziningi, eyethulwe ithimba le-Tencent Hunyuan ngokubambisana ne-Tsinghua University AIR kanye ne-Southeast University. Yakha izindawo ezibonakalayo zokwenziwa ngokuphelele ezisekelwe ku-Honor of Kings, QQ Music kanye ne-Tencent Meeting, iqukethe imisebenzi engu-323 yokushintsha isimo kanye nedatha engaphezu kuka-7.6 万. Ngokusebenzisa ukungalingani okubili kwegebe lolwazi kanye negebe lamathuluzi, ihlola ukuthi amamodeli aqoqa kanjani ulwazi ngokuzimela futhi ahlele izingcingo zamathuluzi ezinyathelo eziningi. Isebenzisa umehluko oqinisekile wesimo sesizindalwazi ukuze ikale, okwenza imiphumela izinze futhi iphinde ikhiqizwe. Le bhentshimakhi ihloselwe abacwaningi bama-agent e-AI nabathuthukisi bamamodeli, ibasiza ukubona amamodeli anenani elihle uma kuqhathaniswa nezindleko futhi iqhubekisele phambili ukusungulwa kwamazinga okuhlola ama-agent ajwayelekile.00
Get3W – Inkundla yokuhlanganisa amamodeli e-AI, i-API eyodwa yokushintsha amamodeli ngokukhululekaInkundla yokuthuthukisaI-Get3W – Inkundla yokuhlanganisa amamodeli e-AI, i-API eyodwa yokushintsha amamodeli ngokukhululeka iyinkundla yokuhlanganisa amamodeli e-AI eyenzelwe onjiniyela namabhizinisi. Nge-API eyodwa ehlanganisiwe, ixhuma amamodeli angaphezu kuka-700 avela kubahlinzeki abangaphezu kuka-20, okuhlanganisa i-Google, OpenAI, ByteDance, Alibaba, Kling, Midjourney, Runway, MiniMax ne-Black Forest Labs. Amakhono ayo ahlanganisa ukukhiqizwa kwamavidiyo, izithombe, umsindo, i-3D, izithombe ze-panoramic nezingxoxo. Onjiniyela badinga ukuhlanganisa kanye kuphela, bese beshintsha amamodeli ngokukhululeka ngaphansi kwesethi efanayo yezindinganiso ze-API, ngaphandle kokuphinda balungiselele i-API yomhlinzeki ngamunye. Inkundla inikeza i-REST API, ama-SDK e-Python/JavaScript, imisebenzi engavumelani nesikhathi kanye nama-callback e-Webhook. Isebenzisa uhlelo lokukhokha ngokuya ngokusetshenziswa, futhi ingonga kufika cishe ku-50% uma iqhathaniswa nokuxhuma ngqo eziteshini ezisemthethweni.00
EpsillaInkundla yokuthuthukisaI-Epsilla iyinkundla ehlanganisa konke yokwakha ama-ejenti e-AI aqhutshwa yidatha eyimfihlo. Kulula ukuyisebenzisa futhi ilungele ochwepheshe bomkhakha; iyalungiseka kakhulu futhi ilungele ochwepheshe be-AI. Ngasikhathi sinye, le nkundla inokuvikeleka okuphelele, iyakhula kalula futhi inamandla okuhlanganiswa nezinye izinhlelo ukuze ihlangabezane nezidingo zamakhasimende ezinkampani.00Wonke amathuluzi akulesi sigaba ayaboniswa.