进阶 www.newapi.ai 2026-11-23 02:52:54 · 0 阅读
第3章 New API Plugins - Model Adapters and Generation Tasks
Plugin marketplaceAdd models and features to your New API gateway.RecommendedNewIncho music generation (prompt-to-song, multilingual vocals)OfficialAll plugins · 57Typed decisions with TypeSafe Jev, input priced per million tokens and output freeOfficialAlibaba Cloud Bailian image and video generation (Wan, Qwen-Image, Z-Image)OfficialSets the reasoning effort from the size of each request; under Where it runs, choose models that support reasoning, as other models may refuse the requestOfficialChecks user text with an OpenAI-compatible moderation API and rejects flagged requestsOfficialVolcengine Doubao Seedance video generation and Seedream image generationOfficialGoogle Veo video generation on the Gemini API (text-to-video and image-to-video)OfficialMiniMax Hailuo video generation (text-to-video, image-to-video, and MiniMax-H3 multimodal reference)OfficialCompresses large images in requests before they are sent upstreamOfficialVolcengine Jimeng video generation (text-to-video, image-to-video, and first-and-last-frame)OfficialKuaishou Kling video generation (text-to-video and image-to-video)OfficialApplies parameter override rules to all requests or to chosen groups, with the same syntax as a channel's parameter override. To change a single channel, use that channel's own parameter override; on the same parameter, the channel's setting winsOfficialRejects requests whose text matches a keyword or regular expressionOfficialReplaces personal information and keys in request text before it is sentOfficialOpenAI Sora video generation (text-to-video, image-to-video, and remix)OfficialSSunoAPI project music and lyrics generationOfficialGoogle Veo and Gemini Omni Flash video generation on Vertex AI (text-to-video and image-to-video)OfficialShengshu Vidu video generation (text-to-video, image-to-video, first-and-last-frame, and reference-to-video)OfficialxAI Grok Imagine video generation (text-to-video, image-to-video, reference-to-video, video editing, and video extension)OfficialMiniMax text, speech, transcription, H3 video, H3-Image and music on one channel@ByAidenWOne web search API for several search vendors (Zhipu, DeepSeek), in the Perplexity Search API format@feitianbubuGemini API on one channel: text, embeddings, images, Omni video, Decision@ByAidenText-to-image and image-to-image with workflows on a self-hosted ComfyUI@seefs001Blocks or masks API keys, tokens, private keys and passwords in requests and model answers@ByAidenBlocks, masks or logs prompts and replies that match administrator keyword lists and regular expressions, by category@ByAidenMuse image generation and editing@gorquanOpenAI-compatible relay that reads upstream balances and disables or restores keys by threshold@ByAidenLuma Ray 3.2 video (text-to-video, image-to-video, keyframes, extension, editing, and reframing) and uni-1 image generation and editing@ByAidenVolcengine visual intelligence synchronous APIs (multilingual OCR)@feitianbubuAlibaba Cloud Bailian text, image, video and speech models on one channel@ByAidenVeo 3.1 and Gemini Omni Flash video generation (text-to-video and image-to-video) on Vertex AI with a service account@ByAidenGrok Imagine image generation and editing@gorquanVolcengine Ark text, embedding, Seedream image and Seedance video models on one channel@ByAidenTencent Hunyuan video generation (HY-Video 1.5, text and image to video) and Hunyuan 3D generation (HY-3D 3.0, 3.1 and Express, text, image and multi-view to 3D) through Tencent Cloud TokenHub@ByAidenELimits which endpoint types a model can use, matched by exact name, keyword or regular expression@visduoAArtbox Artdance video generation with Doubao-compatible client APIs@RexCoding-YuRun fal.ai model endpoints through fal's queue API, with OpenAI-compatible image and video entry points for popular image and video models@ByAidenHHeyGen avatar videos on the Avatar III, IV and V engines (avatar looks and talking photos driven by a script or audio), video translation and lipsync@ByAidenRerank endpoint for self-hosted Ollama models with token billing@ByAidenPixVerse video generation (V6 Max, C1, V6, V5.6, V5.5, V5): text-to-video, image-to-video and first/last frame transitions, with image upload@ByAidenTripo 3D generation (text, image and multiview to 3D on the H and P Series, image to Gaussian splat) and model processing (texture, format conversion, auto rigging, animation retargeting, retopology)@ByAidenFLUX image generation and editing by Black Forest Labs: FLUX 3 Image, FLUX.2 (max, pro, flex, klein) and FLUX.1 (Kontext, 1.1 pro, Ultra, dev, Fill)@ByAidenRun Replicate models as asynchronous predictions: curated official image and video models on the OpenAI image and video APIs, and any other model through a mapped name billed by run time, outputs or tokens@ByAidenRunway Gen-4.5 and Gen-4 Turbo video, Aleph 2.0 video editing, Act-Two character performance, and Gen-4 Image generation@ByAidenESends error metadata to WeCom, Feishu, Telegram or a custom webhook@visduoOllama Cloud web search, web fetch, and Clef System One decisions over native JSON APIs@xuu1998xAI Grok text, Grok Imagine images and video, and speech to text on one channel@ByAidenElevenLabs text to speech, music and sound effects on the OpenAI-compatible speech endpoint@ByAidenWeb search, news, page reading and maps for AI apps, priced per Search1API credit@fatwang2ePhone AI async tasks for video, image, music and speech generation@RixAPIAAIS Service image generation integration plugin@visduoBRun Beeuni Agent tasks and retrieve their text results@demomagicDeprecated: use the Web Search plugin (feitianbubu_search), which serves the same Zhipu engines and more vendors@feitianbubuAAIS Service object storage integration plugin@visduoAAIS Service video generation integration plugin@visduoServe the models running in the Bionic desktop app for chat, Responses, Claude Messages, Gemini, completions and embeddings@rtx233KEnforces the Kimi series request contract: parameter limits, media formats, tool rules and reasoning effort@visduoMeta Model API on one channel: Muse Wonder, Spark and Ember text, Muse Image, SAM 3.1 segmentation and Muse Voice Transcribe@ByAidenCatalog from QuantumNous/new-api-plugins·Updated Nov 22, 2026, 6:50 PM
上一篇
第2章 New API Open-Source AI Gateway: Model Routing, Keys and Usage
下一篇
第4章 AstrLink AI Privacy Gateway - On-Device Redaction for AI Agents