Thola amathuluzi, izindaba, nezifundo
Indawo eyodwa yokuhlola konke okuqukethwe kwe-AI okushicilelwe.
“Codex”Imiphumela ehlobene engu-105
Amathuluzi e-AI
23
Ukuhlela nge-AIIsilinganiso 9.3CodexI-Codex iyi-ejenti yokuhlela nge-AI ye-OpenAI eyenzelwe imisebenzi yobunjiniyela besofthiwe. Imibhalo esemthethweni okwamanje ihlanganisa i-Codex CLI, izandiso ze-IDE, indawo yokusebenza yedeskithophu kanye ne-Codex cloud: i-CLI ingahlola amafayela endaweni yokugcina ikhodi, iguqule ikhodi futhi isebenzise imiyalo; imisebenzi yasefini isebenza ezindaweni ezihlukanisiwe, ingaqhubekiswa ngokuhambisana, futhi iqalwe ku-GitHub, GitLab, Linear noma Slack; abasebenzisi bangabuka izifinyezo nomehluko, bengeze izinguquko noma bakhe izicelo zokuhlanganisa. I-Codex ilungele ukuqonda izinqolobane zamakhodi, ukuthuthukisa izici, ukulungisa amaphutha, ukuhlola nokubuyekeza, kodwa ayithathi indawo yokuhlolwa kwephrojekthi, ukulawulwa kwezimvume kanye nenqubo yokuhlanganisa eyenziwa abantu. Lapho kuhileleka izinqolobane zangempela, okhiye kanye nemisebenzi exhumeke ku-inthanethi, kufanele kuqale kuchazwe ngokucacile ububanzi bokufinyelela.
Ama-API amamodeli amakhuluIsilinganiso 7.5deepkeyI-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.
Ukuklama kwe-UI/UXIsilinganiso 8.4TypeUII-TypeUI inakekelwa yi-Bergside LLC, futhi iyinkundla ehlinzeka ngamathuluzi okubhala ikhodi nge-AI ngomongo wokuklama iphinde ihlaziye ukusebenza kwamawebhusayithi. Abasebenzisi bangaphequlula amakhono okuklama neziqondiso ze-UI ngengosi yewebhu, bakhe noma bangenise izinhlelo zokuklama, bese bexhuma amathuluzi afana ne-Codex, Claude ne-Cursor nge-MCP. Bangasebenzisa ne-Insights ukubuka idatha yethrafikhi, uhambo lwezivakashi, ukuguqulwa nokusebenzisana kwesixhumi esibonakalayo.
Amathuluzi Okuhlela nge-AIIsilinganiso 7.8ConductorI-Conductor iyithuluzi lokubhala ikhodi nge-AI elakhiwe yi-MeltyLabs labathuthukisi be-Mac, elisetshenziselwa ukuhlela ama-agent amaningi okubhala ikhodi e-Claude Code noma e-OpenAI Codex endaweni efanayo yokuthuthukisa. Abasebenzisi bangaqalisa i-agent ezimele yomsebenzi ngamunye, kuyilapho ithuluzi lisebenzisa ama-Git worktree ukuhlukanisa izindawo zokusebenza futhi linikeze isixhumi esibonakalayo esisodwa sokubuka isimo semisebenzi, ukubuyekeza umehluko wekhodi nokusingatha izicelo zokuhlanganisa. Lifanele onjiniyela be-full-stack, abaholi bezobuchwepheshe nabathuthukisi abazimele abadinga ukuqhubekisela phambili amagatsha ezici eziningi ngesikhathi esisodwa, ukuthuthukisa kumarepo amaningi noma ukuqhathanisa izindlela ezahlukene zokuqalisa. I-Conductor imahhala, futhi indawo yokusebenza yekhodi igcinwa ku-Mac yendawo; kodwa ngaphambi kokuyisebenzisa kusadingeka kube nesimo sokungena noma ukhiye we-API we-agent efanele, kanye nokujwayela inqubo yokubuyekeza nokuhlanganisa ehambisana namagatsha asebenza ngokufanayo.
I-ejenti ye-AIOrchard – Uhlaka lwe-Agentic AI oluvulelekile olwakhiwe yi-Microsoft ResearchI-Orchard – uhlaka lwe-Agentic AI oluvulelekile olwakhiwe yi-Microsoft Research. Uhlaka lugxile kusevisi yezindawo i-Orchard Env esekelwe ku-Kubernetes, futhi lusekela ukusetshenziswa kabusha kwama-sandbox ezizindeni ezahlukene ukuze kwenziwe i-data distillation, i-reinforcement learning rollout nokuhlola. Luhlanganisa izimo ezintathu ezinkulu: ubunjiniyela besoftware, ukuzulazula ku-browser kanye nomsizi womuntu siqu. Lunikeza ama-recipe aphelele okuqeqesha e-SFT + RL, kanye nama-trajectory angu-107K e-SWE nedatha engu-3,070 ye-GUI multimodal rollout, futhi lungakwazi ukuqeqeshwa ngokuqondile kusukela ekuqaleni kuye ekugcineni kuma-harness angempela afana ne-Codex ne-OpenClaw. Lubhekiswe emaqenjini ezemfundo nezimboni acwaninga futhi athuthukisa ama-Agent, i-Orchard inezindleko eziphansi zekhompyutha nezokusingathwa, futhi ikhono layo lokusebenza kuma-harness ahlukahlukene lisiza amamodeli amancane ukusondela ekusebenzeni kwezinhlelo ezihamba phambili.
I-ejenti ye-AII-Warp Agent CLI – ithuluzi elizimele lokubhala ikhodi ngomugqa womyalo elivela kwa-WarpI-Warp Agent CLI iyithuluzi elizimele lokubhala ikhodi ngomugqa womyalo elivela kwa-Warp, elisebenza kunoma iyiphi indawo yetheminali, okuhlanganisa i-Ghostty, i-iTerm 2, i-VS Code kanye netheminali eyakhelwe ohlelweni. Leli thuluzi lakhiwe phezu kwengqalasizinda yetheminali ye-Warp, linemultiplexer eyakhelwe ngaphakathi ephethe ukuxhumana kwe-pty ngokwemvelo, futhi ligcina umongo we-Agent lapho ushintsha phakathi kwezinkomba noma kumaseshini e-SSH. Lingasebenzisa izinhlelo ezisebenzisanayo zesikrini esigcwele ezifana ne-sqlite ne-python, lisekele iphrothokholi ye-MCP, futhi lihlele ngokuzenzakalelayo imisebenzi yama-Agent angaphansi asebenza endaweni noma efwini, okuhlanganisa ukuyinikezela ku-Claude Code noma ku-Codex.Izindaba ze-AI
37
Amabucwaningo omkhiqizo2026-08-24Isevisi ye-One Key MCP yethulwe yi-Aliyun Cloud: Ihambisana ne-Qoder, Codex nokunye, futhi ibiza amasevisi amaningi e-MCP ngokuchofoza okukodwaAbathuthukisi bangasebenzisa i-Aliyun Bailian API Key eyodwa ukubiza wonke amasevisi e-MCP ozakwethu be-ecosystem, ihambisana nama-Coding Agent ajwayelekile afana ne-Qoder, Codex, Claude Code ne-Cursor, futhi yenza lula izinqubo zokuxhuma, ukuqinisekisa ubunikazi nokuphatha ukukhokhiswa kwamasevisi amaningi e-MCP.
Amabucwaningo omkhiqizo2026-09-02I-Kimi API manje isekela ngokwemvelo i-Codex ne-Claude CodeI-Moonshot AI imemezele ukuthi i-Kimi API manje isekela amafomethi e-OpenAI Responses API ne-Anthropic Messages API. Abasebenzisi bangaxhuma ngqo kumamodeli ochungechunge lwe-Kimi, okuhlanganisa i-kimi-k3 ne-k2.7, ngaphakathi kwe-Codex ne-Claude Code ngaphandle kokuguqula ifomethi noma ukusebenzisa i-proxy. #Kimi# #AI编程#
Amabucwaningo omkhiqizo2026-08-24I-OpenAI Codex izosetha kabusha isabelo kusasa, iphinde ilungise inkinga yokusetshenziswa ngokushesha esanda kuvelaUmholi wethimba le-OpenAI Codex ne-ChatGPT uphendule ngempendulo yabasebenzisi mayelana nokusetshenziswa kwesabelo ngokushesha, waqinisekisa ukuthi kunezinto ezintathu ezithinta ukubalwa kokukhokhiswa, futhi sekusungulwe ithimba elikhethekile lokuphenya. Izindaba ezinhle ukuthi inkampani ihlela ukusetha kabusha izabelo zabo bonke ababhalisile kusasa, iphinde ikhiphe izinyathelo zokulungisa. #OpenAI#
Amabucwaningo omkhiqizo2026-08-22Akusagcini nje ngokuba okhiye be-API: Ikhono le-GPT‑5.6 Sol le-context yamathokheni ayisigidi ku-Codex selingasetshenziswa nge-akhawunti ye-ChatGPTOnjiniyela be-OpenAI bamemezele ukuthi iwindi le-context lamathokheni ayisigidi lemodeli ye-GPT-5.6 Sol ku-Codex selisekela ama-akhawunti e-ChatGPT. Ababhalisile bangalivula ngokulilungiselela, futhi lilungele ukulungiswa kabusha kwamakhodi amakhulu. Kodwa qaphela ukuthi lizosebenzisa isabelo ngokushesha; kunconywa ukulivula kuphela lapho kudingeka. #OpenAI# #Codex#
Ama-modeli kanye nobuchwepheshe2026-08-22Iphonsela inselelo i-Codex namanye, i-Meta yethula ithuluzi layo lokuqala le-AI agent lokuhlela uhlelo i-Muse CodeInkampani i-Meta namuhla (6 Agasti) ishicilele umbhalo webhulogi, yamemezela ukwethulwa kwenguqulo yokuhlola yethuluzi layo lokuqala le-AI agent lokuhlela uhlelo i-Muse Code, inethemba lokuphonsela inselelo amathuluzi e-Programing Agent afana ne-Claude Code yakwa-Anthropic kanye ne-Codex yakwa-OpenAI.Isifundo se-AI
45
Isifundo se-AI2026-09-08Yakha ama-PCB Alungele Uku-oda nge-PCB Skill ye-Claude Code ne-CodexFunda ukufaka nokusebenzisa i-PCB Skill ne-Claude Code noma i-Codex, i-EasyEDA Pro, kanye ne-automation yesiphequluli. Lesi sifundo sichaza inqubo yayo enezigaba zokuchaza izidingo, ukuthola izingxenye, ukudala nokurutha ibhodi, ukuqinisekisa ngokuzimele amafayela okukhiqiza, nokulungiselela ama-oda ngaphandle kokuvumela i-agent ukuthi ithenge noma yini.
Isifundo se-AI2026-09-01Yakha Indawo Yokusebenza Ye-Codex CLI Egxile Ekusebenzeni Nge-Portable Setup KitFunda ukumisa i-Codex CLI Portable Setup Kit ku-Windows, uqalise ukuhamba komsebenzi kwe-TURBO, uqinisekise ukufakwa komsebenzi owodwa, usebenzise amathuluzi ahlanganisiwe okuhlaziya kabusha, uvumelanise ama-agent e-AI asekelwayo, futhi ubuyisele amafayela angaphambilini ngokubuyisela emuva okusekelwe kumarekhodi okufakwa.
Isifundo se-AI2026-08-31Ukusebenzisa i-MiniMax H3 Cloud ku-Codex ukukhiqiza amavidiyo nge-GPU esefwiniLesi sifundo sichaza indlela yokufaka nokusebenzisa i-MiniMax H3 Cloud for Codex, usebenzisa i-GPU esefwini ye-Youyun Smart Computing ukukhiqiza amavidiyo ngombhalo, ngamafreyimu abalulekile nangezithombe eziyizinkomba, kanye nokuphatha ngokuphepha iziqinisekiso, izindleko nokuvala izimo zokusebenza.
Isifundo se-AI2026-08-28Sebenzisa i-ChatGPT Ukuhlela Nokubuyekeza Imisebenzi Yokubhala Ikhodi ye-CodexFunda indlela yokufaka i-Codex nge-ChatGPT, uyixhume nebhuloho layo le-MCP elivikelwe yi-OAuth, bese usebenzisa i-ChatGPT ukuhlela nokubuyekeza, kuyilapho i-Codex ihlela amafayela futhi iqhuba izivivinyo. Lesi sifundo sichaza nemingcele yezokuphepha yephrojekthi, imiyalo yabathuthukisi, amathuluzi okuxazulula izinkinga, namathiphu athuthukile okusebenza.
Isifundo se-AI2026-08-25Ukwenza ividiyo echaza nge-avatar yedijithali usebenzisa i-Codex Skill: kusukela ekuqaleni komsebenzi kuya ekuhlolweni kwekhwalithiLesi sifundo sichaza indlela yokufaka nokusebenzisa i-lanshu-create-ai-presenter-video, usebenzisa i-Codex noma i-Agent ehambisanayo ukuze kuqedwe ukubhala umbhalo, ukuqopha izwi, ukudala i-avatar yedijithali, ukuvumelanisa ukunyakaza komlomo, imibhalo engezansi, ukuhlela, ukunikezela kanye nokuhlola ikhwalithi.
AI ezentuthuko2026-08-23Isinyathelo ngesinyathelo sokufaka i-Codex yedeskithophu nokuyixhumanisa namamodeli amakhulu aseShayinaLesi sihloko sichaza kakhulu indlela yokufaka i-Codex yedeskithophu, ukuyisebenzisa ngaphandle kwe-akhawunti ye-OpenAI noma i-VPN, nokuyixhumanisa ne-DeepSeek yaseShayina