English · Français · Español · Deutsch · Italiano · Português (Brasil) · Русский · 简体中文 · 日本語 · العربية · हिंदी · বাংলা · اردو · मराठी · తెలుగు · தமிழ் · Kiswahili · Hausa · Tagalog · فارسی (ایران) · አማርኛ · ไทย · Basa Jawa · Bahasa Indonesia · Polski · Tiếng Việt · 繁體中文 · Čeština · 한국어 · Català · ગુજરાતી · ಕನ್ನಡ · Yorùbá · မြန်မာ · Lingála · ଓଡ଼ିଆ · Українська · മലയാളം · ਪੰਜਾਬੀ (ਗੁਰਮੁਖੀ) · Nederlands · Naijíriá Píjin · مصري · پنجابی (شاہ مکھی) · Português (Portugal) · 粵語 · 吴语（上海话） · عربي شامي · عربي سوداني · भोजपुरी · 晋语 · 閩南語 · دارجة جزائرية · 客家話 · الدارجة المغربية · 湘语 · سنڌي · Basa Sunda · Ìgbò · دری · नेपाली · Bahasa Melayu · O‘zbekcha (shimoliy) · isiZulu · عربي صعيدي · پښتو (شمالي) · Oromoo (Giddugaleessa-Lixaa) · سرائیکی · Soomaali · Română · অসমীয়া · Sinugbuanon · Қазақша · සිංහල · isiXhosa · ខ្មែរ · Afrikaans · मैथिली · Wolof · Kurmancî · Ikinyarwanda
Kuna bayyana burin ku. Novera yana shirya albarkatun, aiwatar da aikin kuma yana tsara sakamakon.
Mataimakin AI mai zaman kansa ta TRON-ABS, daga farawa na farko zuwa sakamako: rubutu, hotuna, kiɗa da bidiyo a cikin tattaunawa ɗaya.
Akwai don Windows, Linux, macOS da Android · 80 harsunan sadarwa · 42 masu samar da girgije · 49 saitattun kafofin watsa labarai · Apache 2.0
Kayan aiki da kai · Dashboard · Samfuran AI suna gudana cikin gida akan na'urarka · Tsarukan & Hub · Gajimare · Français · Taimakawa
Tuntuɓar hukuma: contact.novera.tronabs@gmail.com
Bayanan martaba na jama'a na Novera na hukuma: Facebook · Instagram · TikTok · YouTube · X · Bluesky
Raba abubuwan ƙirƙira, ƙirarku, saitunanku, tambayoyinku, ra'ayoyinku da ra'ayoyinku.
> Taimakon fassarar ta kayan aikin fassara, bisa ga Turanci README mai kwanan wata 2026-11-03. Ana maraba da bita na yaren asali. Takardun da aka haɗa suna riƙe ainihin yaren su.
---
## samuwa — 2026-11-03
**Novera 1.0.0 yana samuwa don Windows, Linux, macOS da Android.** [Windows x64 / Linux x64](https://github.com/TRON-ABS/Novera/releases/tag/v1.0.0) · [Windows ARM64](https://github.com/TRON-ABS/Novera/releases/tag/v1.0.0-windows-arm64). Fakitin Desktop, duka CLI ma'ajiyar bayanai, sa hannu da sabunta bayanan jama'a ne kuma an tabbatar dasu.
| Dandalin | samuwa |
| --- | --- |
| Windows x64 | Mai sakawa NSIS na Jama'a, mai sabunta sa hannu; An tabbatar da shigarwa na asali da maye gurbinsu. |
| Linux x64 | Jama'a AppImage, sabunta hannu; An tabbatar da shigarwa na asali da maye gurbinsu. |
| Windows x64 / Linux x64 CLI | [Novera CLI](https://github.com/TRON-ABS/Novera/releases/tag/v1.0.0-cli-novera): `novera.exe` / `novera`, tare da sadaukarwar tashar sabunta sa hannu. |
| Windows ARM64 | [Mai sakawa NSIS na Jama'a](https://github.com/TRON-ABS/Novera/releases/tag/v1.0.0-windows-arm64), Injin rubutu na asali na CPU; an tabbatar da shigarwa, maye gurbin da tashoshi na jama'a. |
| Android ARM64 | [Jama'a sun sanya hannu APK 1.0.0](https://github.com/TRON-ABS/Novera/releases/tag/v1.0.0-android-1000012), Android 7+, tsoho tsari 512 Mai kula ya tabbatar da gina 1000012 yana aiki akan wayar su. Tsaftace fakitin ba tare da tattaunawa ba, samfuri ko maɓallan API na sirri. [Shaida da iyakoki](docs/RELEASE_ANDROID_1.0.0.md). |
| macOS Intel / Apple Silicon | [Sakin 1.0.0 DMG na hukuma](https://github.com/TRON-ABS/Novera/releases/tag/v1.0.0-macos), macOS 13+ Intel / 14+ Apple Silicon; Novera-sa hannun sabuntawa. Ba a sanar da Apple ba; cak na atomatik sun wuce daban akan gine-ginen biyu, an jinkirta duban gani. [Shigarwa da shaida](docs/RELEASE_MACOS_1.0.0.md). |
| iOS | Gina na'urar kwaikwayo ta ciki; Ana buƙatar sa hannu / samarwa Apple kafin shigarwar iPhone. [Bibiya](docs/REPRISE_MACOS_IOS.md). |
| Windows x64 mai šaukuwa | [Sa hannu kan rumbun adana bayanai na ZIP](https://github.com/TRON-ABS/Novera/releases/tag/v1.0.0-windows-portable); bayanai sun tsaya a cikin Data, an tabbatar da maye gurbin Program. |
| Yanar Gizo, MSI, Debian | Abokin ciniki na Yanar Gizo da aka shirya da sauran tsare-tsare sun kasance na ciki; Pages yana samuwa wurin gabatarwa. |
A kan Mac, buɗe DMG kuma kwafi Novera cikin Applications. Idan an katange ƙaddamarwar farko, ba da izinin shiga Novera **Saitunan Tsari → Sirri & Tsaro → Buɗe Ko ta yaya**. [umarnin Apple](https://support.apple.com/guide/mac-help/mh40616/mac).
Sa hannun Sabuntawa ba Windows Authenticode ba ko Apple notarization. Takaddun sigar da hannu ta yanzu ta wuce tashar jama'a; canji tsakanin nau'ikan jama'a guda biyu ya kasance gwajin sakewa na gaba. [Shaida da iyakoki](docs/LAUNCH_STATUS.md).
A kan Android, ana ba da shawarar wayar zamani tare da isasshen ƙwaƙwalwar ajiya don ƙirar gida; yi amfani da mai bada API akan na'urori marasa ƙarfi. Don binciken yanar gizo, mahallin aƙalla **Alamu 16K (16,384)** , akai-akai a cikin nau'ikan iri.
**Windows x64 mai ɗaukar hoto 1.0.0 yana samuwa.** Tattaunawa, saituna da samfuran da aka zazzage suna tsayawa a cikin babban fayil ɗin šaukuwa a ƙarƙashinsa `Data`. Cire cikakken tarihin kuma fara `Novera.exe`. Rufe Novera kafin matsar da babban fayil ko fitar da ajiya USB. [Jagora mai ɗaukar hoto da iyakoki](docs/WINDOWS_PORTABLE.md).
## Dashboard ɗaya don kowane halitta
Novera girma tare da ku. Misalan ku da saitunanku na iya taimaka wa sauran masu amfani samun saitin da ya dace. Raba tambayoyi, batutuwa da ra'ayoyi akan mu [Facebook page](https://www.facebook.com/profile.php?id=61596114515071) ko ta [imel](mailto:contact.novera.tronabs@gmail.com). Idan zai yiwu, haɗa da sigar Novera ɗinku, na'urar da ƙirar gida/API. Cire keɓaɓɓen bayaninka kuma kar a taɓa raba maɓallan API ko kalmomin shiga.
*Hoton hoton aikace-aikacen Windows na ainihi wanda mai kulawa ya kawo akan 1 Nuwamba 2026, wanda ya ba da rahoton cewa aikace-aikacen yana aiki da kyau. Yana nuna shimfidar wuri da aka ƙirƙira, sarrafa fayil ɗin ajiyar sa da haɗewar dashboard ɗin ƙirƙira; wannan ba izgili ba ne.*
Panel ɗaya, hanyoyi biyar: **Atomatik / Rubutu / Hoto / Kiɗa / Bidiyo**. Sarrafa suna bin yanayin da aka zaɓa: rabo, ƙuduri, tsawon lokaci, ƙaranci, inganci, kayan aiki, waƙoƙi, iri da yawa. Faɗin panel da maɓallan sun dace da taga da alamun da aka fassara.
- **1–10 tsararraki masu zuwa.**, batawa zuwa 1, tare da adadi masu zaman kansu don hoto, kiɗa da bidiyo. Tsaida soke aikin mai aiki da sauran jerin gwano.
- **Bazuwar, gyarawa, +1 da -1 tsaba**, saiti ko ƙimar al'ada da zaɓi **Real-ESRGAN ×2/×4** haɓakawa tare da nuna girman ƙarshe.
- Samfurin da aka zaɓa, ƙwaƙwalwar atomatik, haɗe-haɗe, umarni, binciken gidan yanar gizo da kayan aikin suna tsayawa a mashaya iri ɗaya. Kayan aiki sun dogara da damar samfurin kuma an ba da damar shiga.
## Daga farawa na farko zuwa sakamako
**Kayan aikin mu yana haɗa shirye-shiryen zuwa cikakken aikin aiwatar da aiwatarwa.** Novera yana ɗaukar matakan fasaha da ke akwai don daidaitawar da aka zaɓa, yayin da ake samun damar sarrafawa na ci gaba.
| Mataki | Abin da Novera ke sarrafa |
| --- | --- |
| **1. Fara** | Yana zaɓar harshen tsarin da aka goyan baya a farkon ƙaddamarwa kuma yana adana abin da kuke so. Yana gabatar da samfura masu farawa da bayanan kayan masarufi masu amfani. |
| **2. Shirya** | Shigarwa mai jagora yana ba da rubutu, abubuwan sakawa, hoto, kiɗa, bidiyo da haɓakawa. Yana samfotin abubuwan da aka gyara, girma da sarari kyauta kafin amincewar ku, sannan yana bin abubuwan zazzagewa tare da ci gaba, dakatarwa/ci gaba da tabbatarwa SHA-256. |
| **3. Sanya** | Suna yin rajistar samfura da abubuwan haɗinsu, suna shirya bayanan martaba masu dacewa da rahotannin abubuwan dogaro da ake buƙata. Abubuwan da aka dawo da su suna farawa akan CPU tare da adana zaɓin CPU/GPU. Sabbin bayanan martaba na kiɗa suna tsoho zuwa yanayin kayan aiki. |
| **4. Fahimtar bukatar** | Hanyoyin mota sun gane buƙatun ƙirƙirar don rubutu, hoto, kiɗa ko bidiyo. Hakanan zaka iya zaɓar yanayin fitarwa a sarari da ƙirar gida ko gajimare. |
| **5. Kashe** | Ƙwaƙwalwar ajiya ta atomatik tana jiran aiki mai aiki, tana fitar da samfuran da ba a buƙata ba kuma suna sake loda albarkatun da ake buƙata. Jerin ayyuka na 1–10 yana gudana cikin tsari kuma yana fitar da albarkatu kafin aiki na gaba. Masu samar da gajimare suna amfani da zaɓaɓɓun asusun da daidaitawa. |
| **6. Dawo da sake amfani** | Ajiye kafofin watsa labarai a cikin manyan fayilolin tsarin Novera, yana nuna sakamako da lokutan da ake da su, sannan yana ba da bambance-bambance, ɗakin karatu, kwafi, fitarwa ko raba tsarin. Aika zuwa wani aikace-aikacen ya rage aikinku. |
Automation yana rufe injunan haɗaɗɗiya da ayyukan aiki; baya sanya kowane samfurin Hub aiwatarwa. Hardware, abubuwan da aka shigar da kuma samun damar mai bada suna tantance abin da zai iya gudana. Fayiloli, canje-canjen fayil da umarnin harsashi suna bin tallafin samun damar tattaunawar.
[Jagorar mai amfani](docs/USER_GUIDE.md) · [Injiniya da inganci](docs/SUIVI_MOTEURS.md) · [Ci gaba](docs/DEVELOPMENT.md)
A cikin sabbin kafofin, **Saitunan suna daidaita shawarwarinsa zuwa wayar hannu da ƙuntataccen kayan aiki** LFM3-1.7B Q4, ko LFM3-Omni-2B Q4 tare da majiginsa lokacin da rahoton ƙwaƙwalwar ajiya da injin buƙatun sun cika. Ƙarshen gida da ba a samuwa yana ɓoye ko an toshe shi; saita hoton gajimare, kiɗa da bidiyo suna nan. Jagorar mu'amala yana sake yin amfani da kas ɗin yare na 80. Waɗannan ƙididdigar har yanzu suna buƙatar gwajin na'urar da sake ginawa don isa ga shigar da aikace-aikacen.
## Abubuwan iyawa a cikin tushen
Binciken gyare-gyare na zaɓi yanzu ya bambanta ginin gini da gazawar kayan aiki da keɓancewar JavaScript, tare da izini daban-daban da filayen fasaha masu tacewa. An saita faɗakarwar imel na samarwa na sa'a don mai kulawa. Taswirorin tushe masu zaman kansu da cikakkun bayanan faɗakarwa suna bi [kiyaye runbook](docs/MAINTENANCE.md); An tabbatar da rasidin imel na gwaji, yayin da ingantacciyar alamar loda da gwaji a cikin binary na gaba ya rage cak ɗin sakin.
| Yanki | Siffofin |
| --- | --- |
| **Rubutu da fahimtar hoto** | Tattaunawa, haɗe-haɗen daftarin aiki, amsoshi masu gudana da ƙirar gida ko gajimare. Hangen nesa ya dogara da samfurin; fahimta da samar da hotuna dama daban-daban ne. |
| **Hoto** | In-Tattaunawa, tsaba, rabo, ƙuduri, GGUF/Safetensors abubuwan gyara, da aka saita Diffusers babban fayil da goyan bayan NVFP4 CUDA profile. |
| **Kida** | Ƙirƙirar gida ko girgije, tsawon lokaci, inganci, kayan aiki ko muryoyin murya inda aka goyan bayan, tsararrun waƙoƙi da mai kunna sauti. Zaɓuɓɓukan mai bayarwa sun bambanta. |
| **Bidiyo** | Tsarin gida ko girgije, tsawon lokaci, ƙididdige ƙididdigewa/cadence don bayanan martaba na gida, rabo, ƙuduri da mai kunna bidiyo. |
| **Bincike da kayan aiki** | Binciken Yanar Gizo, Maido da tushen tushen daftarin aiki, MCP, wuraren aiki da kayan aikin gida bisa ga goyan bayan samfuri da samun dama. |
| **Sakamako da rabawa** | Laburaren watsa labarai, bambance-bambancen, fitarwa / fitarwa, takardu da kafofin watsa labarai na asali, bugu / PDF da tsarin raba tsarin tushen dandamali. |
| **Gudanar da albarkatu** | Ƙwaƙwalwar ajiya ta atomatik tana kunna ta tsohuwa, saitunan CPU/GPU don abubuwan da aka goyan bayan da bayanan martaba na hannu lokacin da albarkatu suka ba da izini. |
## Samfuran AI suna gudana cikin gida akan na'urarka
Waɗannan samfuran suna samar da tsarin jagora na yanzu. Kuna iya shigo da wasu samfura masu jituwa ko canza bayanan martaba.
| Matsayi | Model / daidaita abubuwan da aka gyara |
| --- | --- |
| **Rubutu - samfurin farko-gudu** | **Nemotron-3.5-Bolt-E4B**, fayil `Nemotron-3.5-Bolt-E4B-Q4_K_M.gguf`. |
| **Dawo da takarda** | **Qwen3-Embedding-0.6B Q8_0**, **CPU ta tsohuwa**, tare da adana zaɓi na CPU/GPU. |
| **Hoto** | **FLUX.2 Klein 4B Q4_K_M**, **Qwen3-4B Q4_K_M** encoder da hukuma FLUX.2 VAE. |
| **Hoto - zaɓi na kayan aiki** | **FLUX.2 Klein 4B NVFP4**, ta hanyar CUDA na asali lokacin da GPU da muhalli suka dace. |
| **Kida** | **ACE-Step 1.5 Turbo 2B Q4_K_M**, `acestep-5Hz-lm-1.7B-Q8_0.gguf` mai tsarawa, sadaukarwa Qwen mai rikodin kiɗa da VAE. |
| **Bidiyo** | **Wan2.1 T2V 1.3B Q4_K_M**, **UMT5-XXL Q4_K_M** encoder da Wan 2.1 VAE. |
| **Ingantawa** | Zabi **Real-ESRGAN ×2 da ×4**, tare da duban abin dogaro. |
Mai rikodin kiɗan Qwen da ƙirar dawo da Qwen sune **fayiloli daban-daban**, an adana shi daban. Lasisin samfurin ya kasance na masu buga su.
[Bayanan rubutu](web-app/src/constants/text-manifest.json) · [Embeddings](web-app/src/constants/embedding-manifest.json) · [Kida](web-app/src/constants/music-manifest.json) · [Abubuwan haɗin Hotuna/bidiyo](src-tauri/src/core/creation_bundle.rs)
## Masu samar da girgije
**42 bayanan martaba** an haɗa su. **ElevenLabs** yanzu yana fitowa a cikin nunin tare da hanyoyin kiɗa daga Stability AI, fal da Replicate.
| | | |
| --- | --- | --- |
| 
[OpenAI](https://platform.openai.com/docs/models) | 
[Microsoft Azure](https://oai.azure.com/deployments) | 
[Anthropic](https://platform.claude.com/docs/en/models/overview) |
| 
[OpenRouter](https://openrouter.ai/models) | 
[Mistral](https://docs.mistral.ai/models) | 
[Groq](https://console.groq.com/docs/models) |
| 
[xAI](https://docs.x.ai/developers/models) | 
[Google Gemini](https://ai.google.dev/gemini-api/docs/models) | 
[MiniMax](https://platform.minimax.io/docs/api-reference/text-openai-api) |
| 
[Hugging Face](https://huggingface.co/models?pipeline_tag=text-generation&inference_provider=all) | 
[NVIDIA NIM](https://build.nvidia.com/models) | 
[DeepSeek](https://api-docs.deepseek.com/) |
| 
[Kimi · Moonshot AI](https://platform.kimi.ai/docs/models) | 
[Alibaba Cloud · Qwen](https://www.alibabacloud.com/help/en/model-studio/models) | 
[Cerebras](https://inference-docs.cerebras.ai/models/overview) |
| 
[Together AI](https://docs.together.ai/docs/serverless/models) | 
[Fireworks AI](https://docs.fireworks.ai/getting-started/models) | 
[Perplexity](https://docs.perplexity.ai/docs/agent-api/models) |
| 
[Meta AI](https://dev.meta.ai/docs/models) | 
[Cohere](https://docs.cohere.com/docs/models) | 
[Z.ai](https://docs.z.ai/guides/llm/glm-5.3) |
| 
[AI21 Labs](https://docs.ai21.com/docs/jamba-foundation-models) | 
[BytePlus · ByteDance](https://docs.byteplus.com/en/docs/modelark/model-parameter-support) | 
[Baidu Qianfan](https://cloud.baidu.com/doc/qianfan/s/rmh4stp0j) |
| 
[Tencent TokenHub](https://cloud.tencent.com/document/product/1823/130078) | 
[Amazon Bedrock](https://docs.aws.amazon.com/bedrock/latest/userguide/inference-chat-completions.html) | 
[IBM watsonx](https://www.ibm.com/docs/en/watsonx/saas?topic=preview-inferencing-gateway-models) |
| 
[ElevenLabs](https://elevenlabs.io/docs/api-reference/music/compose) | 
[Black Forest Labs](https://docs.bfl.ai/quick_start/generating_images) | 
[Stability AI](https://docs.stability.ai/api-reference) |
| 
[Recraft](https://www.recraft.ai/docs/api-reference/agents.md) | 
[Ideogram](https://developer.ideogram.ai/ideogram-api/api-overview) | 
[Adobe Firefly](https://developer.adobe.com/firefly-services/docs/firefly-api/) |
| 
[Runway](https://docs.dev.runwayml.com/api/) | 
[Luma](https://docs.agents.lumalabs.ai/guides/model/) | 
[LTX (Lightricks)](https://docs.ltx.io/models) |
| 
[fal](https://fal.ai/docs/documentation/model-apis/inference/queue) | 
[Replicate](https://replicate.com/docs/reference/http) | 
[Leonardo.Ai](https://docs.leonardo.ai/reference/creategeneration) |
| 
[Krea](https://www.krea.ai/docs/developers/krea-3/overview) | 
[Vidu](https://platform.vidu.com/docs/model-map) | 
[PixVerse](https://docs.platform.pixverse.ai/changelog-2394996m0) |
### Hoto, kiɗa da bidiyo
**Masu samar da 20 suna fallasa hanyoyin tsara**, tare da **49 saitattu: Hoton 25, bidiyo na 20 da kiɗan 4**. Hanyoyin kiɗa sune **ElevenMusic v2.5**, **Stable Audio 3**, **ACE-Step ta hanyar fal** kuma **Stable Audio 2.5 ta hanyar Replicate**.
| Mai bayarwa | Abubuwan da aka fitar a cikin Novera | Samfuran Lissafi |
| --- | --- | --- |
| [ElevenLabs](https://elevenlabs.io/docs/api-reference/music/compose) | Kida | `ElevenMusic v2.5` |
| [Stability AI](https://docs.stability.ai/api-reference/stable-audio/text-to-audio) | Hoto / Kiɗa | `Stable Audio 3` · `Stable Image Ultra` · `Stable Image Core` |
| [fal](https://fal.ai/models/fal-ai/ace-step/api) | Hoto / Kiɗa / Bidiyo | `ACE-Step · cloud` · `Pika 2.2 · fal` · `Qwen Image 2512 · fal` · `FLUX.2 Max · fal` · `Kling 3 Pro · fal` · `Kling 3 Standard · fal` |
| [Replicate](https://replicate.com/stability-ai/stable-audio-2.5/api/schema) | Hoto / Kiɗa | `Stable Audio 2.5 · Replicate` · `FLUX.2 Max · Replicate` |
| [BytePlus · ByteDance](https://docs.byteplus.com/en/docs/ModelArk/image-generation-api) | Hoto / Bidiyo | `Dola Seedream 5 Pro` · `Dreamina Seedance 2.5` |
| [Adobe Firefly](https://developer.adobe.com/firefly-services/docs/firefly-api/api/) | Hoto / Bidiyo | `Adobe Firefly Image 5` · `Adobe Firefly Video` |
| [Google Gemini](https://ai.google.dev/gemini-api/docs/omni) | Hoto / Bidiyo | `Gemini Omni 1.1 Flash` · `Gemini Nano Banana 2.1` |
| [Krea](https://www.krea.ai/docs/developers/krea-3/overview.md) | Hoto | `Krea 3` |
| [Leonardo.Ai](https://docs.leonardo.ai/docs/lucid-origin) | Hoto | `Lucid Origin` · `Krea 2 Turbo · Leonardo` · `Ideogram 3 · Leonardo` |
| [Vidu](https://platform.vidu.com/docs/api-reference/video-models/vidu-q4-preview/image-to-video) | Bidiyo | `Vidu Q4 Preview` · `Vidu Q3 Pro` · `Vidu Q3 Turbo` |
| [OpenAI](https://developers.openai.com/api/docs/guides/image-generation) | Hoto | `GPT Image 2.5 Sunburst` · `GPT Image 2.5 Flare` |
| [xAI](https://docs.x.ai/developers/model-capabilities/imagine) | Hoto / Bidiyo | `Grok Imagine Image 2.0` · `Grok Imagine Video 1.5` · `Grok Imagine Video 1.5 Lite` |
| [Black Forest Labs](https://docs.bfl.ai/quick_start/generating_images) | Hoto | `FLUX.2 [max]` · `FLUX.2 [pro]` · `FLUX.2 [klein] 9B` |
| [Recraft](https://www.recraft.ai/docs/api-reference/agents.md) | Hoto | `Recraft V4.1` · `Recraft V4.1 Flash` · `Recraft V4.1 Pro` |
| [Ideogram](https://developer.ideogram.ai/ideogram-api/api-overview) | Hoto | `Ideogram 4.5` |
| [Runway](https://docs.dev.runwayml.com/api/) | Bidiyo | `Runway Gen-4.5` |
| [Luma](https://docs.agents.lumalabs.ai/guides/model/) | Hoto / Bidiyo | `Luma Uni-1 Max` · `Luma Uni-1` · `Luma Ray 3.2` |
| [LTX (Lightricks)](https://docs.ltx.io/models) | Bidiyo | `LTX-2.5 Pro` · `LTX-2.5 Fast` |
| [MiniMax](https://platform.minimax.io/docs/api-reference/video-generation-v2-create) | Hoto / Bidiyo | `MiniMax H3 Max` · `MiniMax H3` · `MiniMax H3 Highspeed` · `MiniMax Image-01` |
| [PixVerse](https://docs.platform.pixverse.ai/changelog-2394996m0) | Bidiyo | `PixVerse V6 Max` · `PixVerse V6` |
Nuna rubutu da samfurin hangen nesa ta hanyar mai bayarwa
| Mai bayarwa | Samfuran ID da aka bayar a cikin tushen |
| --- | --- |
| [OpenAI](https://platform.openai.com/docs/models) | `gpt-6.1-sol` · `gpt-6-astra` · `gpt-6-luna` · `gpt-6-sol` |
| [Microsoft Azure](https://oai.azure.com/deployments) | Aiwatar da takamaiman asusu ko ƙofar shiga |
| [Anthropic](https://platform.claude.com/docs/en/models/overview) | `claude-sonnet-5-5` · `claude-opus-5-5` · `claude-fable-5-1` · `claude-haiku-5-5` |
| [OpenRouter](https://openrouter.ai/models) | `moonshotai/kimi-k3.1-pro` · `minimax/minimax-m3.1-flash` · `deepseek/deepseek-v4.1-flash-1017` · `nvidia/nemotron-3.5-super-vl:free` |
| [Mistral](https://docs.mistral.ai/models) | `mistral-large-4` · `mistral-medium-3-5` · `mistral-small-2603` · `ministral-3b-2512` |
| [Groq](https://console.groq.com/docs/models) | `openai/gpt-oss-120b` · `openai/gpt-oss-20b` · `llama-3.3-70b-versatile` · `llama-3.1-8b-instant` |
| [xAI](https://docs.x.ai/developers/models) | `grok-4.7` |
| [Google Gemini](https://ai.google.dev/gemini-api/docs/models) | `gemini-4-argon` · `gemini-3.8-flash` · `gemini-3.6-flash` · `gemini-3.5-flash-lite` |
| [MiniMax](https://platform.minimax.io/docs/api-reference/text-openai-api) | `MiniMax-M3.1-Flash` · `MiniMax-M3` |
| [Hugging Face](https://huggingface.co/models?pipeline_tag=text-generation&inference_provider=all) | `deepseek-ai/DeepSeek-V4.1-Flash-1017` · `Qwen/Qwen3.8-27B` · `moonshotai/Kimi-K3` · `zai-org/GLM-5.3` |
| [NVIDIA NIM](https://build.nvidia.com/models) | `minimaxai/minimax-m3.1-flash` · `deepseek-ai/deepseek-v4.1-flash` · `nvidia/nemotron-3.5-super-vl-120b-a12b` · `nvidia/nemotron-3.5-lightning-30b-a3b` · `z-ai/glm-5.3` |
| [DeepSeek](https://api-docs.deepseek.com/) | `deepseek-flash` · `deepseek-v4-pro` |
| [Kimi · Moonshot AI](https://platform.kimi.ai/docs/models) | `kimi-k3.1-pro` · `kimi-k3.1-pro-highspeed` · `kimi-k3` |
| [Alibaba Cloud · Qwen](https://www.alibabacloud.com/help/en/model-studio/models) | `qwen3.8-max` · `qwen3.7-plus` · `qwen3.8-flash` |
| [Cerebras](https://inference-docs.cerebras.ai/models/overview) | `qwen-3.8-27b` · `gpt-oss-120b` |
| [Together AI](https://docs.together.ai/docs/serverless/models) | `deepseek-ai/DeepSeek-V4.1-Flash-1017` · `moonshotai/Kimi-K3` · `Qwen/Qwen3.8-Flash` · `Qwen/Qwen3.8-2.4T-A95B` |
| [Fireworks AI](https://docs.fireworks.ai/getting-started/models) | `accounts/fireworks/models/deepseek-v4p1-flash-1017` · `accounts/fireworks/models/kimi-k3` · `accounts/fireworks/models/minimax-m3` |
| [Perplexity](https://docs.perplexity.ai/docs/agent-api/models) | `perplexity/sonar` · `openai/gpt-6.1-sol` · `google/gemini-4-argon` · `perplexity/kimi-k3` |
| [Meta AI](https://dev.meta.ai/docs/models) | `muse-wonder-1.0` · `muse-spark-1.3` |
| [Cohere](https://docs.cohere.com/docs/models) | `command-a-plus-05-2026` |
| [Z.ai](https://docs.z.ai/guides/llm/glm-5.3) | `glm-5.3` · `glm-5.3-flash` · `glm-5.3-flashx` |
| [AI21 Labs](https://docs.ai21.com/docs/jamba-foundation-models) | `jamba-large` · `jamba-mini-2` |
| [BytePlus · ByteDance](https://docs.byteplus.com/en/docs/modelark/model-parameter-support) | `dola-seed-2-1-turbo-260628` |
| [Baidu Qianfan](https://cloud.baidu.com/doc/qianfan/s/rmh4stp0j) | `ernie-5.1` |
| [Tencent TokenHub](https://cloud.tencent.com/document/product/1823/130078) | `hy4-preview` |
| [Amazon Bedrock](https://docs.aws.amazon.com/bedrock/latest/userguide/inference-chat-completions.html) | `openai.gpt-6.1-sol` |
| [IBM watsonx](https://www.ibm.com/docs/en/watsonx/saas?topic=preview-inferencing-gateway-models) | Aiwatar da takamaiman asusu ko ƙofar shiga |
Abubuwan da aka bincika akan tushen akan **29 Oktoba 2026**. Waɗannan sunaye suna bayyana saitattun Novera, ba garantin samun dama ga kowane asusu ba. Littattafai masu nisa, yankuna, lasisi, ƙayyadaddun ƙididdiga da lissafin kuɗi sun kasance ƙayyadaddun masu bada sabis. Bayanan martaba masu jituwa na iya sabunta kasidarsu; Azure da wasu ƙofofin suna amfani da tura asusu. Kwangilolin Cloud suna da gwaje-gwajen simulators; ba a yi wani tsarar da aka biya don wannan sabuntawar nunin ba.
[Kasidar Media](web-app/src/constants/mediaProviders.ts) · [Samfuran rubutu/hangen nesa](web-app/src/constants/cloudModels.ts) · [Saitin girgije](docs/USER_GUIDE.md#cloud) · [Alamar alama](docs/DEVELOPMENT.md#logos)
## Hub da samfura masu jituwa
**GGUF, ma'auni kuma ƙididdige Safetensors, ONNX, da MLX**: bincika samfura ta ɗawainiya kuma zaɓi bambance-bambancen da suka dace da injin da kayan aikin ku.
| Iyali | Tallafi a cikin tushe | Bukatun |
| --- | --- | --- |
| **GGUF** | Ƙirƙirar rubutu ta hanyar llama.cpp; hotuna da abubuwan bidiyo don bayanan martaba masu goyan baya. | Gine-gine da injin da aka haɗe ya gane. |
| **Raba GGUF** | Bambance-bambancen kowane cikakken jerin, girman da aka tara, zazzage duk-shard da girman/SHA-256/ duban kai. | Daidaitattun sunayen fayil masu lamba; zazzage 500 GB baya nufin samfurin ya dace da wannan PC. |
| **Daidaitaccen Safetensors** | Cikakkun manyan fayiloli ta hanyar Transformers, rubutun dalili akan CPU ko NVIDIA CUDA. | Kanfigareshan, tokenizer da kowane shard da ake buƙata. |
| **ONNX Runtime** | shigo da gida na ƙididdiga na lamba/bool da shirya fitar da rubutu GenAI. | Ingantaccen cikakken babban fayil; takamaiman aiki-takamaiman preprocessing/bayan aiki da ake buƙata; [saitin da iyaka](onnx-server/README.md). |
| **LoRA/PEFT adaftar** | Shigo da cikakken tushe tare da adaftar gida daban a cikin mai bada Safetensors. | Tushen Transformers wanda ba a ƙididdige shi ba, ainihin ma'auni, ƙididdigar tensor da ƙididdigar ƙwaƙwalwar ajiya; [saitin da iyaka](transformers-server/README.md#local-lora--peft-adapters). |
| **FP8 LLM / ModelOpt** | Kisa na asali na CUDA don fitar da Llama/Qwen2/Qwen3 mai yawa mai jituwa kowane-tensor FP8 fitarwa. | Gine-gine/tambayoyi na tensor da binciken kwaya na gaske; [saitin da iyaka](transformers-server/README.md#modelopt-fp8-and-nvfp4-llms). |
| **NVFP4 LLM / ModelOpt** | Kunshin toshe-16 FP4 tsinkaya ta asali CUDA GEMMs; daban daga hoton NVFP4. | Mai jituwa Blackwell GPU, goyan bayan girma / fitarwa da ainihin binciken kwaya; samfuran gwaji da ba a horar ba da aka gwada akan RTX 5070. |
| **AWQ** | 4-bit GEMM samfurin bincike ba tare da haɗin aiki ba. | NVIDIA CUDA da GPTQModel/Transformers na zaɓi. |
| **GPTQ** | 2/3/4/8-bit masu daidaitawa, tsarin gptq/gptq_v2. | NVIDIA CUDA, GPTQModel da Optimum; ainihin kisa da aka gwada a 4 ragowa. |
| **bitsandbytes** | 4-bit NF4/FP4 ko 8-bit INT8. | NVIDIA CUDA da abin dogaro na zaɓi; gwaje-gwaje na ainihi NF4 da INT8. |
| **EXL3** | Cikakkun manyan fayilolin Safetensors da aka ƙididdige su ta hanyar ExLlamaV3; An gwada ainihin tsara tare da CUDA. | NVIDIA GPU, ware Python 3.12 muhalli da kuma jituwa gine; Bayani na CPU |
| **EXL2** | Mai ɗaukar nauyin ExLlamaV2; yawo, sokewa da saukewa an gwada su akan GPU. | CUDA hankali dole ne a zaba a sarari don tarin software da aka gwada; ana buƙatar cikakken wurin bincike na samfur da samuwa VRAM. |
| **MLX** | Samfura masu jituwa akan Apple Silicon. | MLX engine da dandamali mai goyan baya. |
Hub yanzu yana fil **LFM3-1.7B** azaman shawararta na LiquidAI don daidaitawa masu jituwa, tare da bambance-bambancen Q4_K_M, Q6_K da Q8_0. Samfurin-gudu na farko ya kasance wanda aka jera a tsarin jagorar da ke sama.
Tace ma'aunin nauyi daga **karkashin 1 GB zuwa 500 GB da sama**, sannan a rarraba ta hanyar dacewa da hardware, girman, recency ko zazzagewa. Sakamako suna ɗauka a hankali; Ana amfani da tacewa ga samfuran da aka ɗauko da kuma bambance-bambancen da suka dace. Shawarar ƙwaƙwalwar ajiya tana amfani da samuwan RAM/VRAM, mahallin mahalli da abubuwan haɗin haɗin gwiwa, ba tare da ƙwaƙƙwaran ƙidayar GPU ba.
AWQ 4-bit, GPTQ 4-bit, NF4 da INT8 kowannen rubutu da aka kirkira akan ƙirar RTX 5070 Laptop da aka ƙirƙira ta hanyar layi. Wannan yana tabbatar da lodi da aiwatarwa, ba inganci ko aikin kowane samfuri ba. **Waɗannan ƙarin suna cikin tushen: ana buƙatar sake ginawa da dogaro masu dacewa; wannan sabuntawar README baya haɓaka aikin binary ɗin da aka shigar.**
**An auna akan RTX 5070 Laptop:** Qwen2.5-0.5B-Instruct, mahallin 2,048, ƙaddamar da alamun 272, alamun 128 da aka ƙirƙira, gudu biyar masu dumama: ƙaddamarwa na tsakiya **GGUF Q4_K_M 402.7 tokens/s · EXL3 4 bpw 312.5 · EXL2 4 bpw 146.9**. Babu fa'idar saurin gudu sama da GGUF da aka nuna anan. Ƙididdigar ƙididdigewa / daidaitawa ya bambanta, kuma EXL3 ya kasa gwajin lissafi na asali ɗaya; waɗannan sakamakon ba babban gudu ba ne ko matsayi mai inganci. [Cikakken ma'auni da iyakoki](deliverables/exllama-2026-10-30/validation.md).
**TabbyAPI** zaɓin zaɓi ne, haɗin haɗin uwar garken OpenAI mai jituwa daban, wanda aka kashe ta tsohuwa. An gwada gano samfurin EXL3 mai inganci da yawo. Hanyar ExLlama da ake gudanarwa ta kasance daban. [ExLlama saitin da haɗin TabbyAPI](transformers-server/README.md#exl3-first-exl2-second--nvidia-execution).
Tsohuwar nau'in daidaitawar kayan masarufi yana ba da fifiko ga sabbin abubuwan fitarwa tsakanin daidaito iri ɗaya da matakin ƙwaƙwalwar ajiya. Ba a maye gurbin samfuran da aka shigar ta atomatik.
[Kididdige saitin Safetensors](transformers-server/README.md#quantized-safetensors--source-integration-29-october-2026) · [Tsarukan, fifiko da inganci](docs/SUIVI_MOTEURS.md)
Abubuwan da aka sabunta sun ƙara **abubuwan da aka sarrafa na zaɓi na zaɓi a cikin Saituna don Windows x64**: muhallin Python masu zaman kansu, bayanan bayanan CPU/CUDA da aka lika, kimantawa da tabbatarwa kafin saukewa, tabbatarwa, gyarawa da cirewa. Hanyoyi masu fassara na al'ada suna tsayawa ƙarƙashin zaɓuɓɓukan ci gaba. Injunan GGUF na asali suna kiyaye abubuwan da suka kasance; wayar hannu da gajimare ba sa sauke tebur Python/PyTorch. Uku masu zaman kansu CPU / bayanan fitarwa an yi amfani da su tare da bayanan gwaji na ƙirƙira; Mai kula yana ba da rahoton nasarar amfani da tebur akan 1 Nuwamba 2026; Bayanan martaba GPU da tsabta-PC shigarwa har yanzu suna buƙatar keɓance, ingantaccen ingantaccen bayani. [Bayanan martaba da iyakokin gwaji](docs/REPRISE_COMPOSANTS_GERES_80_LANGUES.md). An jinkirta haɗa tari a cikin EXE.
## 80 mu'amala da harsuna da bambance-bambancen karatu
**Duk catalogs na 80 sun rufe maɓallan Turanci na yanzu na 2,429** a fadin 15 sunaye, gami da menus, daidaitawa, saituna, taimako da sarrafa kafofin watsa labarai. Siffofin jam'i da jagorar mu'amala suna bin harshen da aka zaɓa. Amsa da harsunan murya sun dogara daban akan ƙirar.
Wannan ƙirar fasaha ba ta tabbatar da ingancin harshe ba: **bita na asali da daidaitawar yanki sun kasance wajibi**. Duba lakabin asali da sanannun iyakoki a cikin [jagorar harshe](docs/LANGUAGES.md) kuma [Rahoton bincike](deliverables/interface-audit-2026-10-29/RAPPORT.md).
## Taimakawa mutane, a kowane bangare
Bayan makaranta da ilimi, burinmu shine mu taimaka wa mutane su koyi, magance matsaloli, ƙirƙira da haɓaka ayyuka masu amfani a duk duniya. Novera yana nufin sanya ci gaban AI na yanzu cikin sauƙin fahimta da amfani, cikin harsuna, matakan ƙwarewa da kayan aikin da ake da su.
| Sassan da muke son taimakawa ci gaba | Misalai na gudunmawa mai amfani |
| --- | --- |
| Makarantu, ilimi da koyo na rayuwa | Bayyana ra'ayoyi, daidaita kayan koyo da tallafawa malamai, ɗalibai da iyalai. |
| Bincike, kimiyya da sararin samaniya | Bincika adabi, bayyana hanyoyin, tantance bayanai da raba ilimi. |
| Lafiya, kulawa da tallafin zamantakewa | Sanya bayanai cikin sauƙin fahimta da tallafawa bincike da ƙwararrun ayyukan aiki. |
| Noma, abinci da al'ummar karkara | Raba ilimi mai amfani, tallafawa tsarawa da haɓaka kayan aikin gida masu amfani. |
| Muhalli, yanayi, ruwa da halittu | Bayyana shaida, tallafawa lura da taimakawa al'ummomi tsara ayyuka. |
| Makamashi, masana'antu, gine-gine da kere-kere | Takaddun tallafi, ƙira, kiyayewa, koyan fasaha da haɓaka tsari. |
| Kasuwanci, kasuwanci, kudi da kuma aiki | Taimakawa tsara aiki, fahimtar bayanai, haɓaka ayyuka da koyon ƙwarewar sana'a. |
| Ayyukan jama'a, doka da shiga jama'a | Sanya takardu da matakai cikin sauƙin fahimta da sauƙin samun bayanai. |
| Kwamfuta, tsaro ta yanar gizo, robotics da sadarwa | Taimakawa haɓaka software, fahimtar tsarin da gina kayan aiki masu amfani. |
| Arts, al'adu, kafofin watsa labarai, harsuna da al'adun gargajiya | Taimakawa rubuce-rubuce, fassarar, hotuna, bidiyo, ayyukan ƙirƙira da raba al'adu. |
| Sufuri, dabaru, yawon shakatawa, wasanni da rayuwar al'umma | Taimakawa tare da tsarawa, sadarwa, ilmantarwa da ayyukan gida. |
| Dama, aikin jin kai da rayuwar yau da kullum | Rage shingen harshe da ilimi da tallafawa mutane tare da taimakon aiki. |
Waɗannan su ne jagororin bayar da gudummawa da ci gaba, ba wai da'awar cewa kowane fanni na musamman an riga an kawo ko ingantacce ba. Muna son jagora mai sauƙi don amfanin yau da kullun, bayyanannun zaɓuɓɓukan ci gaba lokacin da ake buƙata, kayan aiki na gaskiya da buƙatun girgije, da haɗin gwiwa tare da mutanen da suka san kowane filin da mahallin gida.
## Matsayin aikin
**Novera 1.0.0 yana samuwa don Windows, Linux, macOS da Android.** Hoton hoton yana nuna aikace-aikacen Windows wanda mai kula da shi ya haɗa a cikin gida. Canje-canjen tushen, gwaje-gwajen da aka yi niyya da ingantaccen kayan aiki an rubuta su daban; wannan baya inganta kowane samfuri, dandamali ko asusun gajimare.
An kunna ginin GitHub Actions na hannu don duk bambance-bambancen rarraba goma sha ɗaya. Bi [gina sakamako da tabbatar da zazzagewar ɗan takara](docs/CI_DISTRIBUTIONS.md); ana yin rikodin samuwa daban don kowane dandamali. ’Yan takarar sun bambanta da sa hannun sa hannu na samarwa. An bayyana buƙatun rarrabawa a cikin [Manufar sakin hukuma](OFFICIAL_RELEASES.md) kuma [shafin saki](https://github.com/TRON-ABS/Novera/releases). Ɗaliban nakasassu masu sabuntawa sun kasance na ciki. Fitowar jama'a na buƙatar ingantaccen tsari da ingantaccen tsarin sabuntawa; [duba sa hannun tebur da sauran buƙatun ƙaddamarwa an rubuta su](docs/LAUNCH_STATUS.md).
## Takardu da gudummawa
- [Cikakken jagorar mai amfani](docs/USER_GUIDE.md), [samfur da bayanin mai amfani da dokokin zinare (Faransa)](docs/REGLES_D_OR.md), [samfuri da injuna](docs/SUIVI_MOTEURS.md).
- [Ci gaba](docs/DEVELOPMENT.md), [jagorar gini da saki](docs/BUILD_AND_RELEASE.md), [sigogin gaba da sabunta shirye-shiryen](docs/FUTURES_VERSIONS.md).
- [dashboard na aikin](TABLEAU_DE_BORD_NOVERA.md), [taswirar hanya](ROADMAP.md), [canji](docs/CHANGELOG.md).
- [Gudunmawa](CONTRIBUTING.md): lamba, gwaje-gwaje, bita na asali a cikin yarukan 80 da ra'ayin kwararar aiki na gaske.
## Identity da lasisi
Novera yana kiyaye ta **TRON-ABS** in [wannan ma'ajiyar](https://github.com/TRON-ABS/Novera). Duba cikin [Hanyar ganewa](BRANDING.md) kuma [Manufar sakin hukuma](OFFICIAL_RELEASES.md).
Novera yana ginu a kai [Jan](https://github.com/janhq/jan), wanda Menlo Research da masu ba da gudummawarsa suka kirkira. Siffofin asali da na ɓangare na uku da kuma [lasisin Apache 2.0](LICENSE) ana ajiyewa. Zazzagewar Jan ba fitowar Novera bane.
## Taimakawa Novera
A matsayina na mai bincike mai zaman kansa, na san yadda rashin kuɗi zai iya sa ya zama mai wahala don neman bincike.
Idan kuna son kasancewa cikin kasada ta Novera, zaku iya tallafawa bincike na da ci gaban aikin tare da gudummawar bitcoin na son rai.
Godiya ga kowa da kowa don goyon bayanku, gudunmawarku, ra'ayoyin ku da ƙarfafawa. Kowane motsi yana ƙidaya a cikin wannan kasada!
**Bitcoin (BTC) — cibiyar sadarwar Bitcoin**
Adireshin karba: `17TXpB1aBj5DxjwEfM3K44zyqhixRzY78i`
Don ba da gudummawa, duba wannan lambar QR tare da walat ɗin ku ko kwafi adireshin da ke sama. Zaɓi **Bitcoin (BTC), akan hanyar sadarwa ta Bitcoin**, sai ku zabi duk adadin da kuke so.
---
Ga kowa da kowa ya fara tafiya da Novera: bari sha'awarku ta buɗe sabbin kofofi, ra'ayoyinku su rayu kuma kowace halitta ta ɗauke ku kaɗan kaɗan.
TRON-ABS
Mr Sachet Jean-Marc - Mai bincike mai zaman kansa - Faransa
Fasahar Mutum tare da AI
Godiya ga duk wanda ke ba da gudummawa ga kyakkyawar duniya. Ta hanyar bincikensu, ayyukansu, taimakon juna da ilimin da suke bayarwa, sun taimaka mana mu kai ga inda muke a yau.
Muna yi musu fatan alheri a nan gaba, tare da duk wanda yake dauke da wannan ruhi na rabawa da taimakon wasu gaba.
Kayan aikin haɓakawa: 🧠 Codex - GPT-6.1 Sol