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
Sampeyan nggambarake tujuan sampeyan. Novera nyiapake sumber daya, nglakokake tugas lan ngatur asil.
Asisten AI independen dening TRON-ABS, saka wiwitan nganti asil: teks, gambar, musik lan video ing siji obrolan.
Kasedhiya kanggo Windows, Linux, macOS lan Android · Basa antarmuka 80 · Panyedhiya awan 42 · Prasetel media 49 · Apache 2.0
Otomatisasi · Dashboard · Model AI mlaku sacara lokal ing piranti sampeyan · Format & Hub · Awan · Français · Kontribusi
Kontak resmi: contact.novera.tronabs@gmail.com
Profil sosial resmi Novera: Facebook · Instagram · TikTok · YouTube · X · Bluesky
Nuduhake kreasi, model, setelan, pitakonan, umpan balik, lan ide.
> Terjemahan dibantu piranti terjemahan, adhedhasar README Inggris tanggal 2026-11-03. Review basa asli ditampa. Dokumen sing digandhengake tetep basa asline.
---
## Kasedhiyan — 2026-11-03
**Novera 1.0.0 kasedhiya kanggo Windows, Linux, macOS lan 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). Paket desktop, arsip CLI, teken lan manifes nganyari wis umum lan diverifikasi.
| Platform | Kasedhiyan |
| --- | --- |
| Windows x64 | Pemasang umum NSIS, nganyari nganyari; instalasi native lan panggantos diverifikasi. |
| Linux x64 | Public AppImage, nganyari nganyari; instalasi native lan panggantos diverifikasi. |
| Windows x64 / Linux x64 CLI | [Novera CLI](https://github.com/TRON-ABS/Novera/releases/tag/v1.0.0-cli-novera): `novera.exe` / `novera`, kanthi saluran nganyari sing ditandatangani khusus. |
| Windows ARM64 | [Pemasang umum NSIS](https://github.com/TRON-ABS/Novera/releases/tag/v1.0.0-windows-arm64), mesin teks asli CPU; instalasi, panggantos lan mriksa saluran umum diverifikasi. |
| Android ARM64 | [Ditandatangani umum APK 1.0.0](https://github.com/TRON-ABS/Novera/releases/tag/v1.0.0-android-1000012), Android 7+, kumpulan standar 512. Penyelenggara negesake manawa 1000012 bisa digunakake ing telpon. Paket resik tanpa obrolan, model utawa tombol API pribadi. [Bukti lan watesan](docs/RELEASE_ANDROID_1.0.0.md). |
| macOS Intel / Apple Silicon | [Rilis resmi 1.0.0 DMG](https://github.com/TRON-ABS/Novera/releases/tag/v1.0.0-macos), macOS 13+ Intel / 14+ Apple Silicon; Nganyari Novera. Ora Apple notarized; mriksa otomatis liwati kapisah ing loro arsitektur, kir visual ditundha. [Instalasi lan bukti](docs/RELEASE_MACOS_1.0.0.md). |
| iOS | mbangun simulator internal; Apple tondo asto / provisioning dibutuhake sadurunge instalasi iPhone. [Tindak lanjut](docs/REPRISE_MACOS_IOS.md). |
| Windows x64 portabel | [Arsip ZIP mlebu](https://github.com/TRON-ABS/Novera/releases/tag/v1.0.0-windows-portable); data tetep ing Data, mlebu Program panggantos diverifikasi. |
| Web, MSI, Debian | Klien Web sing di-host lan format liyane tetep internal; Situs presentasi Pages kasedhiya. |
Ing Mac, bukak DMG banjur salin Novera menyang Aplikasi. Yen peluncuran pisanan diblokir, ngidini Novera mlebu **Setelan Sistem → Privasi & Keamanan → Bukak wae**. [Instruksi Apple](https://support.apple.com/guide/mac-help/mh40616/mac).
Tandha nganyari dudu Windows Authenticode utawa notarisasi Apple. Priksa versi saiki manual liwat saluran umum; transisi antarane rong versi umum tetep test sabanjuré-release. [Bukti lan watesan](docs/LAUNCH_STATUS.md).
Ing Android, telpon modern kanthi memori cukup dianjurake kanggo model lokal; gunakake panyedhiya API ing piranti sing kurang kuat. Kanggo panelusuran web, konteks paling ora **Token 16K (16,384)** dianjurake, terus-terusan ing versi.
**Windows x64 portabel 1.0.0 kasedhiya.** Diskusi, setelan lan model sing diundhuh tetep ana ing folder portabel ing ngisor `Data`. Extract arsip lengkap lan miwiti `Novera.exe`. Nutup Novera sadurunge mindhah folder utawa ngeculake panyimpenan USB. [Pandhuan lan watesan portabel](docs/WINDOWS_PORTABLE.md).
## Siji dasbor kanggo saben kreasi
Novera tuwuh karo sampeyan. Conto lan setelan sampeyan bisa mbantu pangguna liyane nemokake persiyapan sing cocog. Nuduhake pitakon, masalah lan ide babagan kita [Facebook page](https://www.facebook.com/profile.php?id=61596114515071) utawa dening [email](mailto:contact.novera.tronabs@gmail.com). Yen bisa, kalebu versi Novera, piranti lan model lokal / API. Mbusak informasi pribadhi lan aja nuduhake kunci utawa sandhi API.
*Tangkapan layar aplikasi Windows nyata sing diwenehake dening maintainer ing 1 November 2026, sing nglaporake manawa aplikasi kasebut bisa digunakake kanthi apik. Iki nuduhake lanskap sing digawe, kontrol file sing disimpen lan dashboard nggawe gabungan; iki dudu mockup.*
Siji panel, limang mode: **Otomatis / Teks / Gambar / Musik / Video**. Kontrol tindakake mode sing dipilih: rasio, resolusi, durasi, irama, kualitas, instrumental, lirik, wiji lan jumlah. Jembar panel lan tombol adaptasi menyang jendhela lan label terjemahan.
- **1–10 generasi sekuensial**, defaulting 1, kanthi jumlah independen kanggo gambar, musik lan video. Stop mbatalake tugas aktif lan antrian sing isih ana.
- **Acak, tetep, + 1 lan −1 wiji**, prasetel utawa rasio khusus lan opsional **Real-ESRGAN ×2/×4** upscaling kanthi dimensi pungkasan ditampilake.
- Model sing dipilih, memori otomatis, lampiran, instruksi, telusuran web lan piranti tetep ana ing garis sing padha. Piranti gumantung saka kapabilitas model lan akses sing diwenehake.
## Saka peluncuran pisanan nganti asil
**Otomatisasi kita nyambungake persiapan menyang alur kerja eksekusi lengkap.** Novera nangani langkah-langkah teknis sing kasedhiya kanggo konfigurasi sing dipilih, nanging tetep bisa diakses kontrol lanjut.
| Langkah | Apa sing ditangani Novera |
| --- | --- |
| **1. Miwiti** | Milih basa sistem sing didhukung nalika diluncurake lan nyimpen pilihan sampeyan. Presents model wiwitan lan informasi hardware migunani. |
| **2. Siapke** | Instalasi sing dipandu nawakake teks, embeddings, gambar, musik, video lan upscaling. Pratinjau komponen, ukuran lan ruang kosong sadurunge disetujoni, banjur nglacak download kanthi kemajuan, ngaso / resume lan verifikasi SHA-256. |
| **3. Ngatur** | Ndhaptar model lan komponene, nyiapake profil sing relevan lan laporan dependensi sing dibutuhake. Embeddings Retrieval diwiwiti ing CPU karo pilihan CPU/GPU disimpen. Profil musik anyar minangka standar kanggo mode instrumental. |
| **4. Ngerti panjaluke** | Rute otomatis ngenali panjalukan nggawe teks, gambar, musik utawa video. Sampeyan uga bisa kanthi tegas milih mode output lan model lokal utawa awan. |
| **5. eksekusi** | Memori otomatis ngenteni kerja aktif, ngeculake model sing ora dibutuhake lan ngisi sumber daya sing dibutuhake. Seri proyek 1–10 mlaku kanthi urutan lan ngeculake sumber daya sadurunge proyek sabanjure. Panyedhiya awan nggunakake akun lan konfigurasi sing dipilih. |
| **6. Nompo lan gunakake maneh** | Nyimpen media ing folder sistem Novera, nampilake asil lan wektu sing kasedhiya, banjur nawakake varian, perpustakaan, nyalin, ekspor utawa enggo bareng sistem. Ngirim menyang aplikasi liyane tetep tumindak sampeyan. |
Otomasi nyakup mesin lan alur kerja terpadu; ora nggawe saben model Hub bisa dieksekusi. Hardware, komponen sing diinstal lan akses panyedhiya nemtokake apa sing bisa ditindakake. Folder, owah-owahan file lan perintah cangkang tindakake wewenang akses obrolan.
[Pandhuan pangguna](docs/USER_GUIDE.md) · [Mesin lan validasi](docs/SUIVI_MOTEURS.md) · [Pangembangan](docs/DEVELOPMENT.md)
Ing sumber sing dianyari, **Konfigurasi nyesuekake rekomendasi kasebut menyang piranti seluler lan piranti keras**: LFM3-1.7B Q4, utawa LFM3-Omni-2B Q4 karo proyektor nalika dilapurake memori lan syarat mesin wis ketemu. Generasi lokal sing ora kasedhiya didhelikake utawa diblokir; gambar maya diatur, musik lan video tetep kasedhiya. Panuntun antarmuka nggunakake maneh katalog basa 80. Perkiraan iki isih mbutuhake tes piranti lan mbangun maneh kanggo nggayuh aplikasi sing wis diinstal.
## Kapabilitas ing sumber
Diagnostik pangopènan opsional saiki mbédakaké kegagalan mbangun lan piranti sing ditangani uga pangecualian JavaScript, kanthi idin sing kapisah lan lapangan teknis sing disaring. Tandha email produksi saben jam dikonfigurasi kanggo maintainer. Peta sumber pribadi lan rincian tandha tindakake ing [buku pedoman pangopènan](docs/MAINTENANCE.md); panrimo email test wis dikonfirmasi, nalika upload simbol asli lan testing ing binar sabanjuré tetep mriksa release.
| Area | Fitur |
| --- | --- |
| **Pangerten teks lan gambar** | Obrolan, lampiran dokumen, balesan streaming lan model lokal utawa awan. Vision gumantung model; pangerten lan ngasilake gambar minangka kemampuan sing kapisah. |
| **Gambar** | Generasi ing obrolan, wiji, rasio, resolusi, komponen GGUF/Safetensors, folder Diffusers sing dikonfigurasi lan profil NVFP4 CUDA asli sing didhukung. |
| **Musik** | Generasi lokal utawa awan, durasi, kualitas, instrumental utawa vokal sing didhukung, lirik terstruktur lan pamuter audio. Pilihan panyedhiya beda-beda. |
| **Video** | Generasi lokal utawa awan, durasi, count pigura/irama kanggo profil lokal, rasio, resolusi lan pamuter video. |
| **Panelusuran lan piranti** | Panelusuran web, njupuk dokumen adhedhasar embedding, MCP, ruang kerja lan alat lokal miturut dhukungan model lan akses sing diwenehake. |
| **Asil lan nuduhake** | Pustaka media, varian, ekspor respon/obrolan, dokumen lan media asli, printing/PDF lan enggo bareng sistem gumantung platform. |
| **Manajemen sumber daya** | Memori otomatis diaktifake kanthi gawan, setelan CPU/GPU kanggo komponen sing didhukung lan profil manual yen sumber daya ngidini. |
## Model AI mlaku sacara lokal ing piranti sampeyan
Model iki mbentuk konfigurasi dipandu saiki. Sampeyan bisa ngimpor model liyane sing kompatibel utawa ngganti profil.
| Peran | Model / komponen sing dikonfigurasi |
| --- | --- |
| **Teks — model pisanan** | **Nemotron-3.5-Bolt-E4B**, berkas `Nemotron-3.5-Bolt-E4B-Q4_K_M.gguf`. |
| **Pengambilan dokumen** | **Qwen3-Embedding-0.6B Q8_0**, **CPU minangka standar**, kanthi pilihan CPU/GPU sing disimpen. |
| **Gambar** | **FLUX.2 Klein 4B Q4_K_M**, **Qwen3-4B Q4_K_M** enkoder lan resmi FLUX.2 VAE. |
| **Gambar — opsi hardware** | **FLUX.2 Klein 4B NVFP4**, liwat jalur CUDA asli nalika GPU lan lingkungan kompatibel. |
| **Musik** | **ACE-Step 1.5 Turbo 2B Q4_K_M**, `acestep-5Hz-lm-1.7B-Q8_0.gguf` planner, encoder musik Qwen khusus lan VAE. |
| **Video** | **Wan2.1 T2V 1.3B Q4_K_M**, **UMT5-XXL Q4_K_M** enkoder lan Wan 2.1 VAE. |
| **Upscaling** | Opsional **Real-ESRGAN ×2 lan ×4**, kanthi mriksa ketergantungan. |
Encoder musik Qwen lan model retrieval Qwen yaiku **macem-macem file**, disimpen kanthi kapisah. Lisensi model tetep dadi penerbit.
[Manifes teks](web-app/src/constants/text-manifest.json) · [Embeddings](web-app/src/constants/embedding-manifest.json) · [Musik](web-app/src/constants/music-manifest.json) · [Komponen gambar/video](src-tauri/src/core/creation_bundle.rs)
## Panyedhiya awan
**Profil panyedhiya 42** terintegrasi. **ElevenLabs** saiki katon ing showcase bebarengan karo rute musik saka Stability AI, fal lan 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) |
### Gambar, musik lan video
**Panyedhiya 20 mbukak rute generasi**, karo **prasetel 49: gambar 25, video 20 lan musik 4**. Rute musik yaiku **ElevenMusic v2.5**, **Stable Audio 3**, **ACE-Step nganti fal** lan **Stable Audio 2.5 nganti Replicate**.
| Panyedhiya | Output ing Novera | Model katalog |
| --- | --- | --- |
| [ElevenLabs](https://elevenlabs.io/docs/api-reference/music/compose) | Musik | `ElevenMusic v2.5` |
| [Stability AI](https://docs.stability.ai/api-reference/stable-audio/text-to-audio) | Gambar / Musik | `Stable Audio 3` · `Stable Image Ultra` · `Stable Image Core` |
| [fal](https://fal.ai/models/fal-ai/ace-step/api) | Gambar / Musik / Video | `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) | Gambar / Musik | `Stable Audio 2.5 · Replicate` · `FLUX.2 Max · Replicate` |
| [BytePlus · ByteDance](https://docs.byteplus.com/en/docs/ModelArk/image-generation-api) | Gambar / Video | `Dola Seedream 5 Pro` · `Dreamina Seedance 2.5` |
| [Adobe Firefly](https://developer.adobe.com/firefly-services/docs/firefly-api/api/) | Gambar / Video | `Adobe Firefly Image 5` · `Adobe Firefly Video` |
| [Google Gemini](https://ai.google.dev/gemini-api/docs/omni) | Gambar / Video | `Gemini Omni 1.1 Flash` · `Gemini Nano Banana 2.1` |
| [Krea](https://www.krea.ai/docs/developers/krea-3/overview.md) | Gambar | `Krea 3` |
| [Leonardo.Ai](https://docs.leonardo.ai/docs/lucid-origin) | Gambar | `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) | Video | `Vidu Q4 Preview` · `Vidu Q3 Pro` · `Vidu Q3 Turbo` |
| [OpenAI](https://developers.openai.com/api/docs/guides/image-generation) | Gambar | `GPT Image 2.5 Sunburst` · `GPT Image 2.5 Flare` |
| [xAI](https://docs.x.ai/developers/model-capabilities/imagine) | Gambar / Video | `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) | Gambar | `FLUX.2 [max]` · `FLUX.2 [pro]` · `FLUX.2 [klein] 9B` |
| [Recraft](https://www.recraft.ai/docs/api-reference/agents.md) | Gambar | `Recraft V4.1` · `Recraft V4.1 Flash` · `Recraft V4.1 Pro` |
| [Ideogram](https://developer.ideogram.ai/ideogram-api/api-overview) | Gambar | `Ideogram 4.5` |
| [Runway](https://docs.dev.runwayml.com/api/) | Video | `Runway Gen-4.5` |
| [Luma](https://docs.agents.lumalabs.ai/guides/model/) | Gambar / Video | `Luma Uni-1 Max` · `Luma Uni-1` · `Luma Ray 3.2` |
| [LTX (Lightricks)](https://docs.ltx.io/models) | Video | `LTX-2.5 Pro` · `LTX-2.5 Fast` |
| [MiniMax](https://platform.minimax.io/docs/api-reference/video-generation-v2-create) | Gambar / Video | `MiniMax H3 Max` · `MiniMax H3` · `MiniMax H3 Highspeed` · `MiniMax Image-01` |
| [PixVerse](https://docs.platform.pixverse.ai/changelog-2394996m0) | Video | `PixVerse V6 Max` · `PixVerse V6` |
Tampilake dhaptar cekak model teks lan visi miturut panyedhiya
| Panyedhiya | ID model sing ditawakake ing sumber |
| --- | --- |
| [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) | Penyebaran utawa gateway khusus akun |
| [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) | Penyebaran utawa gateway khusus akun |
Persediaan dicenthang ing sumber ing **29 Oktober 2026**. Jeneng kasebut nggambarake prasetel Novera, ora dijamin akses kanggo saben akun. Katalog adoh, wilayah, lisensi, kuota lan tagihan tetep khusus panyedhiya. Profil sing kompatibel bisa refresh katalog; Azure lan sawetara gateway nggunakake panyebaran akun. Kontrak awan duwe tes simulasi; ora ana generasi mbayar dileksanakake kanggo nganyari showcase iki.
[Katalog media](web-app/src/constants/mediaProviders.ts) · [Model teks/vision](web-app/src/constants/cloudModels.ts) · [Persiyapan awan](docs/USER_GUIDE.md#cloud) · [Ikon asalipun](docs/DEVELOPMENT.md#logos)
## Hub lan model sing kompatibel
**GGUF, standar lan kuantitatif Safetensors, ONNX, lan MLX**: njelajah model miturut tugas lan pilih varian sing cocog karo mesin lan piranti keras sampeyan.
| Kulawarga | Dhukungan ing sumber | Syarat |
| --- | --- | --- |
| **GGUF** | Nggawe teks liwat llama.cpp; komponen gambar lan video kanggo profil didhukung. | Arsitektur dikenali dening mesin bundled. |
| **Pisah GGUF** | Siji varian saben seri lengkap, ukuran sing dijumlah, download kabeh-shard lan mriksa ukuran/SHA-256/header. | Jeneng berkas standar; ngundhuh 500 GB ora ateges model kasebut cocog karo PC. iki |
| **Standar Safetensors** | Rampungake folder liwat Transformers, teks sebab ing CPU utawa NVIDIA CUDA. | Konfigurasi, tokenizer lan saben beling sing dibutuhake. |
| **ONNX Runtime** | Impor lokal grafik angka/bool lan ekspor teks GenAI sing disiapake. | Folder lengkap sing divalidasi; preprocessing / postprocessing khusus tugas dibutuhake; [persiyapan lan watesan](onnx-server/README.md). |
| **Adaptor LoRA / PEFT** | Impor basis lengkap karo adaptor lokal sing kapisah ing panyedhiya Safetensors. | Dasar Transformers penyebab sing ora dikuantisasi, identitas sing cocog, tes tensor lan perkiraan memori kumulatif; [persiyapan lan watesan](transformers-server/README.md#local-lora--peft-adapters). |
| **FP8 LLM / ModelOpt** | Eksekusi CUDA asli kanggo ekspor Llama/Qwen2/Qwen3 per-tensor FP8 sing kompatibel. | Pemeriksa arsitektur/tensor lan probe kernel nyata; [persiyapan lan watesan](transformers-server/README.md#modelopt-fp8-and-nvfp4-llms). |
| **NVFP4 LLM / ModelOpt** | Dikemas blok-16 FP4 proyeksi liwat asli CUDA GEMM; kapisah saka gambar NVFP4. | Kompatibel Blackwell GPU, didhukung dimensi/ekspor lan probe kernel nyata; model uji sing durung dilatih diuji ing RTX 5070. |
| **AWQ** | 4-bit GEMM model checkpoints tanpa operasi fusi. | NVIDIA CUDA lan loader GPTQModel/Transformers opsional. |
| **GPTQ** | Konfigurasi 2/3/4/8-bit, format gptq/gptq_v2. | NVIDIA CUDA, GPTQModel lan Optimum; eksekusi nyata dites ing 4 bit. |
| **bitsandbytes** | 4-bit NF4/FP4 utawa 8-bit INT8. | NVIDIA CUDA lan dependensi opsional; tes NF4 lan INT8 nyata. |
| **EXL3** | Folder model Safetensors kuantitatif lengkap dimuat liwat ExLlamaV3; generasi nyata karo CUDA dites. | NVIDIA GPU, lingkungan Python 3.12 sing kapisah lan arsitektur sing kompatibel; ora CPU fallback. |
| **EXL2** | Loader khusus ExLlamaV2; streaming, pembatalan lan mbongkar dites ing GPU. | CUDA Attention backend kudu dipilih kanthi jelas kanggo tumpukan piranti lunak sing diuji; model checkpoint lengkap lan kasedhiya VRAM dibutuhake. |
| **MLX** | Model sing kompatibel ing Apple Silicon. | Mesin MLX lan platform sing didhukung. |
Hub saiki pin **LFM3-1.7B** minangka rekomendasi LiquidAI kanggo konfigurasi sing kompatibel, kanthi varian Q4_K_M, Q6_K lan Q8_0. Model sing diluncurake pisanan tetep dadi sing kadhaptar ing konfigurasi sing dipandu ing ndhuwur.
Filter bobot download saka **ing 1 GB nganti 500 GB lan ndhuwur**, banjur urut miturut pas hardware, ukuran, kekinian utawa download. Asil dimuat kanthi progresif; saringan ditrapake kanggo model sing dijupuk lan varian standar. Saran memori nggunakake kasedhiya RAM/VRAM, konteks lan komponen kanca, tanpa memori GPU sing dienggo bareng kaping pindho.
AWQ 4-bit, GPTQ 4-bit, NF4 lan INT8 saben ngasilake teks ing RTX 5070 Laptop nggunakake model cilik sing ora dilatih sing digawe offline. Iki validasi loading lan eksekusi, dudu kualitas utawa kinerja saben model. **Tambahan iki ana ing sumber: mbangun maneh lan dependensi sing kompatibel dibutuhake; nganyari README iki ora upgrade binar diinstal.**
**Diukur ing RTX 5070 Laptop:** Qwen2.5-0.5B-Instruct, konteks 2,048, prompt 272 token, 128 ngasilake token, limang warmed run: decode median **GGUF Q4_K_M 402.7 tokens/s · EXL3 4 bpw 312.5 · EXL2 4 bpw 146.9**. Ora entuk manfaat kacepetan liwat GGUF sing dituduhake ing kene. Quantization / kalibrasi beda-beda, lan EXL3 gagal siji tes aritmetika prasaja; asil iki ora kacepetan umum utawa peringkat kualitas. [Pangukuran lan watesan lengkap](deliverables/exllama-2026-10-30/validation.md).
**TabbyAPI** minangka sambungan server sing cocog karo OpenAI opsional, dikelola kanthi kapisah, dipateni kanthi gawan. Panemuan lan streaming model EXL3 sing wis dikonfirmasi dites. Rute ExLlama lokal sing dikelola tetep béda. [persiyapan ExLlama lan sambungan TabbyAPI](transformers-server/README.md#exl3-first-exl2-second--nvidia-execution).
Urut standar sing diadaptasi saka piranti keras luwih milih rilis sing luwih anyar ing tingkat kompatibilitas sing padha lan tingkat memori sing kasedhiya. Model sing diinstal ora diganti kanthi otomatis.
[Persiyapan Safetensors kuantitatif](transformers-server/README.md#quantized-safetensors--source-integration-29-october-2026) · [Format, prioritas lan validasi](docs/SUIVI_MOTEURS.md)
Sumber sing dianyari nambahake **komponen sing dikelola opsional ing Konfigurasi kanggo Windows x64**: lingkungan Python pribadi, pinned CPU/CUDA profil, prakiraan lan konfirmasi sadurunge download, verifikasi, ndandani lan mbusak. Path interpreter adat tetep ing opsi majeng. Mesin GGUF asli njaga komponen sing ana; seluler lan maya ora download desktop Python / PyTorch. Telung profil CPU / ekspor pribadi dileksanakake kanthi data uji sintetik; Penyelenggara laporan sukses nggunakake desktop ing 1 November 2026; Profil GPU lan instalasi resik-PC isih perlu kapisah, validasi rinci. [Profil lan watesan tes](docs/REPRISE_COMPOSANTS_GERES_80_LANGUES.md). Semat tumpukan ing EXE ditundha.
## basa lan varian antarmuka 80
**Kabeh katalog 80 nutupi kunci basa Inggris 2,429 saiki** antarane ruang jeneng 15, kalebu menu, konfigurasi, setelan, bantuan lan kontrol media. Wangun jamak lan arah antarmuka tindakake basa sing dipilih. Basa tanggepan lan swara gumantung ing model kasebut.
Cakupan teknis iki ora nandhakake kualitas linguistik: **review asli lan adaptasi regional tetep perlu**. Waca label native lan watesan dikenal ing [pandhuan basa](docs/LANGUAGES.md) lan [laporan audit](deliverables/interface-audit-2026-10-29/RAPPORT.md).
## Mbantu wong, ing saben sektor
Ngluwihi sekolah lan pendhidhikan, ambisi kita yaiku mbantu wong sinau, ngrampungake masalah, nggawe lan ngembangake proyek sing migunani ing saindenging jagad. Novera duwe tujuan supaya kemajuan AI saiki luwih gampang dimangerteni lan digunakake, ing antarane basa, tingkat pengalaman lan hardware sing kasedhiya.
| Sektor sing arep kita bantu berkembang | Conto kontribusi migunani |
| --- | --- |
| Sekolah, pendidikan lan sinau seumur hidup | Nerangake konsep, adaptasi materi sinau lan ndhukung guru, siswa lan kulawarga. |
| Riset, ilmu lan ruang | Jelajahi literatur, nerangake metode, nganalisa data lan nuduhake kawruh. |
| Kesehatan, perawatan lan dhukungan sosial | Nggawe informasi luwih gampang dimangerteni lan ndhukung riset lan alur kerja profesional. |
| Pertanian, pangan lan masyarakat deso | Nuduhake kawruh praktis, ndhukung perencanaan lan ngembangake alat sing migunani sacara lokal. |
| Lingkungan, iklim, banyu lan keanekaragaman hayati | Nerangake bukti, ndhukung pengamatan lan mbantu komunitas ngrancang proyek. |
| Energi, industri, konstruksi lan kerajinan | Dhukungan dokumentasi, desain, pangopènan, sinau teknis lan perbaikan proses. |
| Bisnis, perdagangan, keuangan lan lapangan kerja | Mbantu ngatur karya, mangerteni informasi, ngembangake proyek lan sinau katrampilan profesional. |
| Layanan umum, hukum lan partisipasi sipil | Nggawe dokumen lan prosedur luwih gampang dimangerteni lan informasi luwih gampang diakses. |
| Komputasi, keamanan siber, robotika lan telekomunikasi | Mbantu ngembangake piranti lunak, ngerti sistem lan mbangun alat sing migunani. |
| Seni, budaya, media, basa lan warisan | Dhukungan nulis, terjemahan, gambar, video, proyek kreatif lan enggo bareng budaya. |
| Transportasi, logistik, pariwisata, olahraga lan urip masyarakat | Bantuan babagan perencanaan, komunikasi, sinau lan inisiatif lokal. |
| Aksesibilitas, karya kamanungsan lan urip saben dina | Ngurangi alangan basa lan kawruh lan ndhukung wong kanthi bantuan praktis. |
Iki minangka pituduh kanggo kontribusi lan pangembangan, dudu pratelan manawa saben aplikasi khusus sektor wis dikirim utawa divalidasi. Kita pengin pandhuan sing gampang kanggo panggunaan saben dina, pilihan sing jelas yen dibutuhake, syarat hardware lan awan sing transparan, lan kolaborasi karo wong sing ngerti saben lapangan lan konteks lokal.
## Status proyek
**Novera 1.0.0 kasedhiya kanggo Windows, Linux, macOS lan Android.** Gambar nuduhake aplikasi Windows sing disusun sacara lokal dening pangurus. Owah-owahan sumber, tes sing ditarget lan validasi hardware didokumentasikake kanthi kapisah; iki ora validasi saben model, platform utawa akun maya.
Manual GitHub Actions mbangun diaktifake kanggo kabeh sewelas varian distribusi. Tindakake [gawe asil lan download calon sing wis diverifikasi](docs/CI_DISTRIBUTIONS.md); kasedhiyan direkam dhewe kanggo saben platform. Calon beda karo rilis produksi sing ditandatangani. Syarat distribusi diterangake ing [kabijakan rilis resmi](OFFICIAL_RELEASES.md) lan [rilis kaca](https://github.com/TRON-ABS/Novera/releases). Kandidat sing ora aktif nganyari saiki tetep internal. Rilis umum mbutuhake mekanisme nganyari sing dikonfigurasi lan diverifikasi; [mriksa teken desktop lan syarat peluncuran isih didokumentasikan](docs/LAUNCH_STATUS.md).
## Dokumentasi lan kontribusi
- [Pandhuan pangguna sing rinci](docs/USER_GUIDE.md), [aturan emas informasi produk lan pangguna (Prancis)](docs/REGLES_D_OR.md), [model lan mesin](docs/SUIVI_MOTEURS.md).
- [Pangembangan](docs/DEVELOPMENT.md), [pandhuan mbangun lan ngeculake](docs/BUILD_AND_RELEASE.md), [versi mbesuk lan persiapan nganyari](docs/FUTURES_VERSIONS.md).
- [Dasbor proyek](TABLEAU_DE_BORD_NOVERA.md), [peta dalan](ROADMAP.md), [changelog](docs/CHANGELOG.md).
- [Kontribusi](CONTRIBUTING.md): kode, tes, review asli ing basa 80 lan umpan balik alur kerja nyata.
## Identitas lan lisensi
Novera dikelola dening **TRON-ABS** ing [gudang iki](https://github.com/TRON-ABS/Novera). Waca [kabijakan identitas](BRANDING.md) lan [kabijakan rilis resmi](OFFICIAL_RELEASES.md).
Novera dibangun ing [Jan](https://github.com/janhq/jan), digawe dening Menlo Research lan kontributor. Atribusi asli lan pihak katelu lan [lisensi Apache 2.0](LICENSE) disimpen. Download Jan dudu rilis Novera.
## Dhukungan Novera
Minangka peneliti independen, aku ngerti kepriye kekurangan dana bisa dadi angel kanggo nerusake riset.
Yen sampeyan pengin dadi bagian saka petualangan Novera, sampeyan bisa ndhukung risetku lan pangembangan proyek kanthi kontribusi bitcoin sukarela.
Matur nuwun kanggo kabeh wong kanggo dhukungan, kontribusi, umpan balik lan dorongan. Saben patrap penting ing petualangan iki!
**Bitcoin (BTC) — jaringan Bitcoin**
Alamat panampa: `17TXpB1aBj5DxjwEfM3K44zyqhixRzY78i`
Kanggo nyumbang, pindai kode QR iki nganggo dompet utawa salin alamat ing ndhuwur. Pilih **Bitcoin (BTC), ing jaringan Bitcoin**, banjur pilih jumlah sing dikarepake.
---
Kanggo kabeh wong sing miwiti lelungan karo Novera: muga-muga rasa penasaran sampeyan mbukak lawang anyar, ide-ide sampeyan urip lan saben ciptaan bakal nggawa sampeyan luwih cepet.
TRON-ABS
Pak Sachet Jean-Marc — Peneliti independen — Prancis
Seni Manungsa nganggo AI
Matur nuwun kanggo kabeh wong sing nyumbang kanggo jagad sing luwih apik. Liwat panemuan, karya, gotong royong lan kawruh sing diwarisake, dheweke wis nulungi kita tekan saiki.
Kita ngarep-arep sing paling apik kanggo masa depan, bebarengan karo kabeh wong sing duwe semangat nuduhake lan mbantu wong liya maju.
Alat pangembang: 🧠 Codex — GPT-6.1 Sol