Omniinfer.io

AI & Agent Cloud for DevelopersShip models and agents in minutes, call 200+ models with one API, and run secure, fast agent sandboxes— developer-first and startup-friendly.Sign UpBook a DemoTRUSTED BYMODEL APISRun 200+ AI models with a simple APIPlug-and-play access to the latest LLMs, image, video, TTS, embeddings models and more via simple APIs. Scale from prototype to production without managing infrastructure.Browse 200+ modelsCUSTOM MODELSEnterprise-Grade Custom Models, Zero Infrastructure HassleDeploy custom models with guaranteed performance SLAs, limitless scalability, and round-the-clock monitoring—no DevOps needed. Focus on innovation, not infrastructure.Launch Your ModelCUSTOM MODELSEnterprise-Grade Custom Models, Zero Infrastructure HassleDeploy custom models with guaranteed performance SLAs, limitless scalability, and round-the-clock monitoring—no DevOps needed. Focus on innovation, not infrastructure.Launch Your ModelAGENT SANDBOXSafe, instant runtimes for AI agentsRun autonomous agents in isolated containers with ~200 ms startups, safe tool use (browser/API/code), and massive concurrency. Billed per-second for CPU/RAM.Explore NowAGENT SANDBOXSafe, instant runtimes for AI agentsRun autonomous agents in isolated containers with ~200 ms startups, safe tool use (browser/API/code), and massive concurrency. Billed per-second for CPU/RAM.Explore NowGPU CLOUDGlobally distributed GPUs with local speedLaunch high-performance GPU instances in seconds across global regions. Ideal for training, finetuning, and high-throughput inference. Use on-demand or spot at 50% off.Explore GPUsGPU CLOUDGlobally distributed GPUs with local speedLaunch high-performance GPU instances in seconds across global regions. Ideal for training, finetuning, and high-throughput inference. Use on-demand or spot at 50% off.Explore GPUsWhy Novita AI01Developer-first DXSimple APIs/SDKs, clear docs, instant scale.02PerformanceHigh throughput LLM serving and low-latency startup for agents.03Global & reliableDeploy close to users on resilient infrastructure.04Cost-efficientSave up to 50% with smart pricing and spot where it fits. 05Bring your own modelPrivate endpoints & custom SLAs when you need them.06Secure by designSandbox isolation for agent workloads.Testimonials"Novita has been instrumental in optimizing our AI workflows at beBee.com, powering over 90% of our token usage with exceptional performance and competitive pricing. Their support is unparalleled—truly 11 out of 10—and far exceeds that of other providers we've worked with. We're excited to continue scaling with Novita."Javier Cámara-Rica, Co-Founder and CEOGo to Case Study"Novita has been a huge help for us at Fish Audio. Their reliable GPU infrastructure allows us focus on developing and improving our text-to-speech models instead of dealing with hardware headaches. Their support and performance have made it much easier to push our work forward."Shijia Liao, Founder and CEO"Novita's Model API was super simple to integrate, and it's been great in powering our AI-driven flashcards and quizzes. The platform takes care of the heavy lifting, so we can focus on building better learning tools for our users without worrying about infrastructure or scaling issues."Petros Christodoulou, Co-Founder and CEO"Working with Novita has completely simplified how we deploy, scale, and host our AI models. Their platform is reliable and efficient, making it easy to manage even complex deployments. They've quickly proven to be a dependable partner we can trust to support our needs!"Wei Zhu, Solution Architect"Novita has been instrumental in optimizing our AI workflows at beBee.com, powering over 90% of our token usage with exceptional performance and competitive pricing. Their support is unparalleled—truly 11 out of 10—and far exceeds that of other providers we've worked with. We're excited to continue scaling with Novita."Javier Cámara-Rica, Co-Founder and CEOGo to Case Study"Novita has been a huge help for us at Fish Audio. Their reliable GPU infrastructure allows us focus on developing and improving our text-to-speech models instead of dealing with hardware headaches. Their support and performance have made it much easier to push our work forward."Shijia Liao, Founder and CEO"Novita's Model API was super simple to integrate, and it's been great in powering our AI-driven flashcards and quizzes. The platform takes care of the heavy lifting, so we can focus on building better learning tools for our users without worrying about infrastructure or scaling issues."Petros Christodoulou, Co-Founder and CEO"Working with Novita has completely simplified how we deploy, scale, and host our AI models. Their platform is reliable and efficient, making it easy to manage even complex deployments. They've quickly proven to be a dependable partner we can trust to support our needs!"Wei Zhu, Solution Architect"Novita has been instrumental in optimizing our AI workflows at beBee.com, powering over 90% of our token usage with exceptional performance and competitive pricing. Their support is unparalleled—truly 11 out of 10—and far exceeds that of other providers we've worked with. We're excited to continue scaling with Novita."Javier Cámara-Rica, Co-Founder and CEOGo to Case Study"Novita has been a huge help for us at Fish Audio. Their reliable GPU infrastructure allows us focus on developing and improving our text-to-speech models instead of dealing with hardware headaches. Their support and performance have made it much easier to push our work forward."Shijia Liao, Founder and CEO"Novita's Model API was super simple to integrate, and it's been great in powering our AI-driven flashcards and quizzes. The platform takes care of the heavy lifting, so we can focus on building better learning tools for our users without worrying about infrastructure or scaling issues."Petros Christodoulou, Co-Founder and CEO"Working with Novita has completely simplified how we deploy, scale, and host our AI models. Their platform is reliable and efficient, making it easy to manage even complex deployments. They've quickly proven to be a dependable partner we can trust to support our needs!"Wei Zhu, Solution Architect"Novita has been instrumental in optimizing our AI workflows at beBee.com, powering over 90% of our token usage with exceptional performance and competitive pricing. Their support is unparalleled—truly 11 out of 10—and far exceeds that of other providers we've worked with. We're excited to continue scaling with Novita."Javier Cámara-Rica, Co-Founder and CEOGo to Case Study"Novita has been a huge help for us at Fish Audio. Their reliable GPU infrastructure allows us focus on developing and improving our text-to-speech models instead of dealing with hardware headaches. Their support and performance have made it much easier to push our work forward."Shijia Liao, Founder and CEO"Novita's Model API was super simple to integrate, and it's been great in powering our AI-driven flashcards and quizzes. The platform takes care of the heavy lifting, so we can focus on building better learning tools for our users without worrying about infrastructure or scaling issues."Petros Christodoulou, Co-Founder and CEO"Working with Novita has completely simplified how we deploy, scale, and host our AI models. Their platform is reliable and efficient, making it easy to manage even complex deployments. They've quickly proven to be a dependable partner we can trust to support our needs!"Wei Zhu, Solution ArchitectWhat's newAnnouncementMiniMax M2.5Available on Novita AI nowLLM$0.3/1.2 in/out MTokens | 204800 ContextAnnouncementGLM-5Available on Novita AI nowLLM$1.0/3.2 in/out MTokens | 202800 ContextAnnouncementQwen3 Coder NextAvailable on Novita AI nowLLM$0.2/1.5 in/out M Tokens | 262144 ContextAnnouncementNovitaXHugging FaceNovita available on Hugging Face nowAnnouncementNovitaXPOENovita models on POE nowAnnouncementNovitaXLLMAccelerate AI InferenceAnnouncementNovitaXAccelerate AI InferenceAnnouncementFeatured BlogsAI insights, LLM tips, and GPU solutionsCHECK OUT THE LATEST ARTICLESWhat's newAnnouncementMiniMax M2.5Available on Novita AI nowLLM$0.3/1.2 in/out MTokens | 204800 ContextAnnouncementGLM-5Available on Novita AI nowLLM$1.0/3.2 in/out MTokens | 202800 ContextAnnouncementQwen3 Coder NextAvailable on Novita AI nowLLM$0.2/1.5 in/out M Tokens | 262144 ContextAnnouncementNovitaXHugging FaceNovita available on Hugging Face nowAnnouncementNovitaXPOENovita models on POE nowAnnouncementNovitaXLLMAccelerate AI InferenceAnnouncementNovitaXAccelerate AI InferenceAnnouncementFeatured BlogsAI insights, LLM tips, and GPU solutionsCHECK OUT THE LATEST ARTICLESReady to build smarter? Start today.Get started with Novita AI and unlock the power of affordable, reliable, and scalable AI inference for your applications.Get StartedBook a Demo --- PricingExplore pricing for our Model APIs and GPU resources. Find the right plan to match your needs with transparent rates and flexible options.PricingExplore pricing for our Model APIs and GPU resources. Find the right plan to match your needs with transparent rates and flexible options.Serverless EndpointsDedicated EndpointsAgent SandboxGPUsBatch inference is available at an introductory 50% discount on input and output tokens for supported models.Learn moreAllLLMImageAudioVideoCacheProviders:AllDeepseekAdvanced AI models from DeepSeek, offering cutting-edge reasoning capabilities and competitive pricing for enterprise and research applications.Model NameContextInputOutputActionsdeepseek/deepseek-v3.2163,840$0.269 /Mt· Cache Read $0.1345 /Mt$0.4 /MtMoredeepseek/deepseek-ocr-28,192$0.03 /Mt$0.03 /MtMoredeepseek/deepseek-v3.2-exp163,840$0.27 /Mt$0.41 /MtMoredeepseek/deepseek-v3.1-terminus131,072$0.27 /Mt· Cache Read $0.135 /Mt$1 /MtMoredeepseek/deepseek-v3.1131,072$0.27 /Mt· Cache Read $0.135 /Mt$1 /MtMoredeepseek/deepseek-v3-0324163,840$0.27 /Mt· Cache Read $0.135 /Mt$1.12 /MtMoredeepseek/deepseek-r1-0528163,840$0.7 /Mt· Cache Read $0.35 /Mt$2.5 /MtMoredeepseek/deepseek-r1-distill-llama-70b8,192$0.8 /Mt$0.8 /MtMoredeepseek/deepseek-prover-v2-671b160,000$0.7 /Mt$2.5 /MtMoredeepseek/deepseek-v3-turbo64,000$0.4 /Mt$1.3 /MtMoredeepseek/deepseek-r1-turbo64,000$0.7 /Mt$2.5 /MtMoreQwenQwen series models offering efficient language processing with various parameter sizes, from lightweight to enterprise-grade solutions.Model NameContextInputOutputActionsqwen/qwen3.5-27b262,144$0.3 /Mt$2.4 /MtMoreqwen/qwen3.5-122b-a10b262,144$0.4 /Mt$3.2 /MtMoreqwen/qwen3.5-35b-a3b262,144$0.25 /Mt$2 /MtMoreqwen/qwen3.5-397b-a17b262,144$0.6 /Mt$3.6 /MtMoreqwen/qwen3-coder-next262,144$0.2 /Mt$1.5 /MtMoreqwen/qwen3-vl-235b-a22b-thinking131,072$0.98 /Mt$3.95 /MtMoreqwen/qwen3-next-80b-a3b-instruct131,072$0.15 /Mt$1.5 /MtMoreqwen/qwen3-next-80b-a3b-thinking131,072$0.15 /Mt$1.5 /MtMoreqwen/qwen3-vl-235b-a22b-instruct131,072$0.3 /Mt$1.5 /MtMoreqwen/qwen3-max262,144-Tiered pricingMoreqwen/qwen3-coder-480b-a35b-instruct262,144$0.3 /Mt$1.3 /MtMoreqwen/qwen3-coder-30b-a3b-instruct160,000$0.07 /Mt$0.27 /MtMoreqwen/qwen3-235b-a22b-thinking-2507131,072$0.3 /Mt$3 /MtMoreqwen/qwen3-235b-a22b-instruct-2507131,072$0.09 /Mt$0.58 /MtMoreqwen/qwen-2.5-72b-instruct32,000$0.38 /Mt$0.4 /MtMoreqwen/qwen3-235b-a22b-fp840,960$0.2 /Mt$0.8 /MtMoreqwen/qwen2.5-vl-72b-instruct32,768$0.8 /Mt$0.8 /MtMoreqwen/qwen3-32b-fp840,960$0.1 /Mt$0.45 /MtMoreqwen/qwen3-30b-a3b-fp840,960$0.09 /Mt$0.45 /MtMoreqwen/qwen3-vl-8b-instruct131,072$0.08 /Mt$0.5 /MtMoreqwen/qwen3-vl-30b-a3b-instruct131,072$0.2 /Mt$0.7 /MtMoreqwen/qwen3-vl-30b-a3b-thinking131,072$0.2 /Mt$1 /MtMoreqwen/qwen3-omni-30b-a3b-thinking65,536-OmnimodalMoreqwen/qwen3-omni-30b-a3b-instruct65,536-OmnimodalMoreqwen/qwen-mt-plus16,384$0.25 /Mt$0.75 /MtMoreqwen/qwen3-4b-fp8128,000$0.03 /Mt$0.03 /MtMoreqwen/qwen2.5-7b-instruct32,000$0.07 /Mt$0.07 /MtMoreBaiduBaidu's ERNIE models providing advanced Chinese language understanding and multimodal capabilities, optimized for Chinese applications with competitive pricing.Model NameContextInputOutputActionsbaidu/ernie-4.5-21B-a3b-thinking131,072$0.07 /Mt$0.28 /MtMorebaidu/ernie-4.5-vl-424b-a47b123,000$0.42 /Mt$1.25 /MtMorebaidu/ernie-4.5-300b-a47b-paddle123,000$0.28 /Mt$1.1 /MtMorebaidu/ernie-4.5-vl-28b-a3b-thinking131,072$0.39 /Mt$0.39 /MtMorebaidu/ernie-4.5-vl-28b-a3b30,000$0.14 /Mt$0.56 /MtMorebaidu/ernie-4.5-21B-a3b120,000$0.07 /Mt$0.28 /MtMoreZai-orgGLM series models from Tsinghua University, featuring advanced Chinese language understanding and generation capabilities.Model NameContextInputOutputActionszai-org/glm-4.7-flash200,000$0.07 /Mt· Cache Read $0.01 /Mt$0.4 /MtMorezai-org/glm-5-turbo202,800$1.2 /Mt· Cache Read $0.24 /Mt$4 /MtMorezai-org/glm-5202,800$1 /Mt· Cache Read $0.2 /Mt$3.2 /MtMorezai-org/glm-4.7204,800$0.6 /Mt· Cache Read $0.11 /Mt$2.2 /MtMorezai-org/autoglm-phone-9b-multilingual65,536$0.035 /Mt$0.138 /MtMorezai-org/glm-4.6v131,072$0.3 /Mt· Cache Read $0.055 /Mt$0.9 /MtMorezai-org/glm-4.6204,800$0.55 /Mt· Cache Read $0.11 /Mt$2.2 /MtMorezai-org/glm-4.5131,072$0.6 /Mt· Cache Read $0.11 /Mt$2.2 /MtMorezai-org/glm-4.5v65,536$0.6 /Mt· Cache Read $0.11 /Mt$1.8 /MtMorezai-org/glm-4.5-air131,072$0.13 /Mt· Cache Read $0.025 /Mt$0.85 /MtMoreSao10KSpecialized fine-tuned models optimized for creative and roleplay applications with enhanced storytelling capabilities.Model NameContextInputOutputActionssao10k/l3-70b-euryale-v2.18,192$1.48 /Mt$1.48 /MtMoresao10k/l3-8b-lunaris8,192$0.05 /Mt$0.05 /MtMoreSao10K/L3-8B-Stheno-v3.28,192$0.05 /Mt$0.05 /MtMoresao10k/l31-70b-euryale-v2.28,192$1.48 /Mt$1.48 /MtMoreMiniMaxMinimax AI's advanced language models delivering robust conversational AI capabilities with optimized performance for customer service, content generation, and creative applications, featuring strong multilingual support and enterprise-ready scalability.Model NameContextInputOutputActionsminimax/minimax-m2.7204,800$0.3 /Mt· Cache Read $0.06 /Mt$1.2 /MtMoreminimax/minimax-m2.5-highspeed204,800$0.6 /Mt· Cache Read $0.03 /Mt$2.4 /MtMoreminimax/minimax-m2.5204,800$0.3 /Mt· Cache Read $0.03 /Mt$1.2 /MtMoreminimax/minimax-m2.1204,800$0.3 /Mt· Cache Read $0.03 /Mt$1.2 /MtMoreminimax/minimax-m2204,800$0.3 /Mt· Cache Read $0.03 /Mt$1.2 /MtMoreminimaxai/minimax-m1-80k1,000,000$0.55 /Mt$2.2 /MtMoreMoonshotAI--Model NameContextInputOutputActionsmoonshotai/kimi-k2.5262,144$0.6 /Mt· Cache Read $0.1 /Mt$3 /MtMoremoonshotai/kimi-k2-thinking262,144$0.6 /Mt· Cache Read $0.15 /Mt$2.5 /MtMoremoonshotai/kimi-k2-0905262,144$0.6 /Mt$2.5 /MtMoremoonshotai/kimi-k2-instruct131,072$0.57 /Mt$2.3 /MtMoreKwaiKAT--Model NameContextInputOutputActionskwaipilot/kat-coder-pro256,000$0.3 /Mt· Cache Read $0.06 /Mt$1.2 /MtMoreOpenAI--Model NameContextInputOutputActionsopenai/gpt-oss-120b131,072$0.05 /Mt$0.25 /MtMoreopenai/gpt-oss-20b131,072$0.04 /Mt$0.15 /MtMoreLlamaMeta's Llama models providing state-of-the-art language understanding with open architecture designed for diverse applications.Model NameContextInputOutputActionsmeta-llama/llama-3.1-8b-instruct16,384$0.02 /Mt$0.05 /MtMoremeta-llama/llama-3.3-70b-instruct131,072$0.135 /Mt$0.4 /MtMoremeta-llama/llama-3-8b-instruct8,192$0.04 /Mt$0.04 /MtMoremeta-llama/llama-3-70b-instruct8,192$0.51 /Mt$0.74 /MtMoremeta-llama/llama-4-maverick-17b-128e-instruct-fp81,048,576$0.27 /Mt$0.85 /MtMoremeta-llama/llama-4-scout-17b-16e-instruct131,072$0.18 /Mt$0.59 /MtMoreMistral--Model NameContextInputOutputActionsmistralai/mistral-nemo60,288$0.04 /Mt$0.17 /MtMoreGemmaGoogle's Gemma models offering high-quality language processing with excellent performance for various NLP tasks.Model NameContextInputOutputActionsgoogle/gemma-3-27b-it98,304$0.119 /Mt$0.2 /MtMoreOOthers--Model NameContextInputOutputActionsxiaomimimo/mimo-v2-flash262,144$0.1 /Mt· Cache Read $0.02 /Mt$0.3 /MtMoremicrosoft/wizardlm-2-8x22b65,535$0.62 /Mt$0.62 /MtMorenousresearch/hermes-2-pro-llama-3-8b8,192$0.14 /Mt$0.14 /MtMoreEmbeddingsModel NameContextInputqwen/qwen3-embedding-0.6b32768$0.07 /Mtqwen/qwen3-embedding-8b32768$0.07 /Mtbaai/bge-m38192$0.01 /MtImagePricing may vary based on image dimensions, inference steps, and upscaling factors. Use thePricing Calculatorfor an estimate.API NameWidth&HeightSteps/ScalePricingText to Image512*5125$0.001 /imageImage to Image512*5125$0.001 /imageRemove Background--$0.017 /imageReplace Background--$0.0255 /imageInpainting512*5125$0.0015 /imageRemove Text--$0.017 /imageCleanup--$0.017 /imageMerge Face--$0.0255 /imageAPI NameModeWidth&HeightPricingFlux.1 Kontext Dev--$0.0225 /imagefast_mode-$0.018 /imageFlux.1 Kontext Max--$0.072 /imageFlux.1 Kontext Pro--$0.036 /imageGLM Image Generation--$0.0143 /imageHunyuan Image 3--$0.1 /imageImage Eraser--$0.0250 /imageImage Remove Background--$0.0180 /imageImage Upscaler--$0.0100 /imageQwen-Image Edit--$0.02 /imageQwen-Image Text to Image--$0.02 /imageSeedream 3.0 Text to Image--$0.03 /imageSeedream 4.0--$0.03 /imageSeedream 4.5--$0.0300 /imageSeedream 5.0 lite--$0.0350 /imageZ Image Turbo--$0.0050 /imageZ Image Turbo LoRA--$0.0100 /imageVideoPricing may vary based on the number of frames, chosen model, and inference steps. Use thePricing Calculatorfor an estimate.API NameTotal FramesStepsPricingText to Video3220$0.0307 /videoAPI NameModeDurationResolutionPricingHunyuan Video Fast-5s1280*720 | 720*1280$0.30 /videoKling V1.6 Image to VideoStandard5s720P$0.27 /videoStandard10s720P$0.54 /videoProfessional5s1080P$0.46 /videoProfessional10s1080P$0.92 /videoKling V1.6 Text to VideoStandard5s720P$0.27 /videoStandard10s720P$0.54 /videoKling V2.5 Turbo Image to Video-5s1080P$0.35 /video-10s1080P$0.70 /videoKling V2.5 Turbo Text to Video-5s1080P$0.35 /video-10s1080P$0.70 /videoKling V2.6 Pro Image to VideoNo Audio5s1080P$0.35 /videoNo Audio10s1080P$0.70 /videoAudio5s1080P$0.70 /videoAudio10s1080P$1.40 /videoKling V2.6 Pro Image-to-VideoNo Audio5s$0.3500 /videoAudio5s$0.7000 /videoNo Audio10s$0.7000 /videoAudio10s$1.4000 /videoKling V2.6 Pro Motion Control--$0.0700 /sKling V2.6 Pro Text to VideoNo Audio5s1080P$0.35 /videoNo Audio10s1080P$0.70 /videoAudio5s1080P$0.70 /videoAudio10s1080P$1.40 /videoKling V2.6 Pro Text-to-VideoNo Audio5s$0.3500 /videoAudio5s$0.7000 /videoNo Audio10s$0.7000 /videoAudio10s$1.4000 /videoKling v3.0 Pro Image-to-VideoNo Audio-$0.224 /sAudio-$0.336 /sKling v3.0 Pro Image-to-VideoAudio-$0.3360 /sNo Audio-$0.2240 /sKling v3.0 Pro Text-to-VideoNo Audio-$0.224 /sAudio-$0.336 /sKling v3.0 Pro Text-to-VideoAudio-$0.3360 /sNo Audio-$0.2240 /sKling v3.0 Standard Image-to-VideoNo Audio-$0.168 /sAudio-$0.252 /sKling v3.0 Standard Image-to-VideoAudio-$0.2520 /sNo Audio-$0.1680 /sKling v3.0 Standard Text-to-VideoNo Audio-$0.168 /sAudio-$0.252 /sKling v3.0 Standard Text-to-VideoNo Audio-$0.1680 /sAudio-$0.2520 /sKling-o1 Image to Video--$0.1120 /sKling-o1 Reference Video GenerationVIDEO+IMAGE-$0.1680 /sONLY VIDEO-$0.1680 /sONLY IMAGE-$0.1680 /sONLY PROMPT-$0.1680 /sKling-o1 Text to Video--$0.1120 /sKling-o1 Video EditingStandard mode-$0.1680 /videoFast mode-$0.0900 /videoMinimax Hailuo 2.3 Fast Image to Video-6s768P$0.19 /video-10s768P$0.32 /video-6s1080P$0.33 /videoMinimax Hailuo 2.3 Image to Video-6s768P$0.28 /video-10s768P$0.56 /video-6s1080P$0.49 /videoMinimax Hailuo 2.3 Text to Video-6s768P$0.28 /video-10s768P$0.56 /video-6s1080P$0.49 /videoMiniMax Video 01-6s720P$0.40 /videoMiniMax Video 02-6s768P$0.25 /video-10s768P$0.50 /video-6s1080P$0.44 /videoPixVerse V4.5 Image to Video-5s360P$0.25 /video-5s540P$0.25 /video-5s720P$0.35 /video-5s1080P$0.70 /videofast_mode5s360P$0.50 /videofast_mode5s540P$0.50 /videofast_mode5s720P$0.70 /videoPixVerse V4.5 Text to Video-5s360P$0.25 /video-5s540P$0.25 /video-5s720P$0.35 /video-5s1080P$0.70 /videofast_mode5s360P$0.50 /videofast_mode5s540P$0.50 /videofast_mode5s720P$0.70 /videoSeedance 1.5 Pro Image To VideoFF/Online/Silent-480p$0.0120 /sFF/Batch/Audio-480p$0.0120 /sFF/Online/Audio-480p$0.0240 /sFF/Batch/Silent-720p$0.0130 /sFF/Online/Silent-720p$0.0260 /sFF/Batch/Audio-720p$0.0260 /sFF/Online/Audio-720p$0.0520 /sFF/Batch/Silent-480p$0.0060 /sFLF/Batch/Silent-480p$0.0060 /sFLF/Online/Silent-480p$0.0120 /sFLF/Batch/Audio-480p$0.0120 /sFLF/Online/Audio-480p$0.0240 /sFLF/Batch/Silent-720p$0.0130 /sFLF/Online/Silent-720p$0.0260 /sFLF/Batch/Audio-720p$0.0260 /sFLF/Online/Audio-720p$0.0520 /sSeedance 1.5 Pro Text To VideoFF/Batch/Silent-480p$0.0060 /sFF/Online/Silent-480p$0.0120 /sFF/Batch/Audio-480p$0.0120 /sFF/Online/Audio-480p$0.0240 /sFF/Batch/Silent-720p$0.0130 /sFF/Online/Silent-720p$0.0260 /sFF/Batch/Audio-720p$0.0260 /sFF/Online/Audio-720p$0.0520 /sSeedance V1 Lite Image to Video-5s480P( 21:9 & 9:21 )$0.08 /video-5s480P( 16:9 & 9:16 )$0.09 /video-5s480P( 4:3 & 3:4 )$0.08 /video-5s480P( 1:1 )$0.09 /video-5s720P( 21:9 & 9:21 )$0.20 /video-5s720P( 16:9 & 9:16 )$0.19 /video-5s720P( 4:3 & 3:4 )$0.20 /video-5s720P( 1:1 )$0.19 /video-5s1080P( 21:9 & 9:21 )$0.43 /video-5s1080P( 16:9 & 9:16 )$0.44 /video-5s1080P( 4:3 & 3:4 )$0.44 /video-5s1080P( 1:1 )$0.44 /video-10s480P( 21:9 & 9:21 )$0.17 /video-10s480P( 16:9 & 9:16 )$0.17 /video-10s480P( 4:3 & 3:4 )$0.17 /video-10s480P( 1:1 )$0.17 /video-10s720P( 21:9 & 9:21 )$0.41 /video-10s720P( 16:9 & 9:16 )$0.37 /video-10s720P( 4:3 & 3:4 )$0.39 /video-10s720P( 1:1 )$0.39 /video-10s1080P( 21:9 & 9:21 )$0.85 /video-10s1080P( 16:9 & 9:16 )$0.88 /video-10s1080P( 4:3 & 3:4 )$0.88 /video-10s1080P( 1:1 )$0.87 /videoSeedance V1 Lite Text to Video-5s480P( 21:9 & 9:21 )$0.08 /video-5s480P( 16:9 & 9:16 )$0.09 /video-5s480P( 4:3 & 3:4 )$0.08 /video-5s480P( 1:1 )$0.09 /video-5s720P( 21:9 & 9:21 )$0.20 /video-5s720P( 16:9 & 9:16 )$0.19 /video-5s720P( 4:3 & 3:4 )$0.20 /video-5s720P( 1:1 )$0.19 /video-5s1080P( 21:9 & 9:21 )$0.43 /video-5s1080P( 16:9 & 9:16 )$0.44 /video-5s1080P( 4:3 & 3:4 )$0.44 /video-5s1080P( 1:1 )$0.44 /video-10s480P( 21:9 & 9:21 )$0.17 /video-10s480P( 16:9 & 9:16 )$0.17 /video-10s480P( 4:3 & 3:4 )$0.17 /video-10s480P( 1:1 )$0.17 /video-10s720P( 21:9 & 9:21 )$0.41 /video-10s720P( 16:9 & 9:16 )$0.37 /video-10s720P( 4:3 & 3:4 )$0.39 /video-10s720P( 1:1 )$0.39 /video-10s1080P( 21:9 & 9:21 )$0.85 /video-10s1080P( 16:9 & 9:16 )$0.88 /video-10s1080P( 4:3 & 3:4 )$0.88 /video-10s1080P( 1:1 )$0.87 /videoSeedance V1 Pro Image to Video-5s480P( 21:9 & 9:21 )$0.12 /video-5s480P( 16:9 & 9:16 )$0.12 /video-5s480P( 4:3 & 3:4 )$0.12 /video-5s480P( 1:1 )$0.12 /video-5s720P( 21:9 & 9:21 )$0.28 /video-5s720P( 16:9 & 9:16 )$0.26 /video-5s720P( 4:3 & 3:4 )$0.27 /video-5s720P( 1:1 )$0.27 /video-5s1080P( 21:9 & 9:21 )$0.59 /video-5s1080P( 16:9 & 9:16 )$0.61 /video-5s1080P( 4:3 & 3:4 )$0.61 /video-5s1080P( 1:1 )$0.61 /video-10s480P( 21:9 & 9:21 )$0.23 /video-10s480P( 16:9 & 9:16 )$0.24 /video-10s480P( 4:3 & 3:4 )$0.23 /video-10s480P( 1:1 )$0.24 /video-10s720P( 21:9 & 9:21 )$0.56 /video-10s720P( 16:9 & 9:16 )$0.51 /video-10s720P( 4:3 & 3:4 )$0.55 /video-10s720P( 1:1 )$0.54 /video-10s1080P( 21:9 & 9:21 )$1.18 /video-10s1080P( 16:9 & 9:16 )$1.22 /video-10s1080P( 4:3 & 3:4 )$1.22 /video-10s1080P( 1:1 )$1.22 /videoSeedance V1 Pro Text to Video-5s480P( 21:9 & 9:21 )$0.12 /video-5s480P( 16:9 & 9:16 )$0.12 /video-5s480P( 4:3 & 3:4 )$0.12 /video-5s480P( 1:1 )$0.12 /video-5s720P( 21:9 & 9:21 )$0.28 /video-5s720P( 16:9 & 9:16 )$0.26 /video-5s720P( 4:3 & 3:4 )$0.27 /video-5s720P( 1:1 )$0.27 /video-5s1080P( 21:9 & 9:21 )$0.59 /video-5s1080P( 16:9 & 9:16 )$0.61 /video-5s1080P( 4:3 & 3:4 )$0.61 /video-5s1080P( 1:1 )$0.61 /video-10s480P( 21:9 & 9:21 )$0.23 /video-10s480P( 16:9 & 9:16 )$0.24 /video-10s480P( 4:3 & 3:4 )$0.23 /video-10s480P( 1:1 )$0.24 /video-10s720P( 21:9 & 9:21 )$0.56 /video-10s720P( 16:9 & 9:16 )$0.51 /video-10s720P( 4:3 & 3:4 )$0.55 /video-10s720P( 1:1 )$0.54 /video-10s1080P( 21:9 & 9:21 )$1.18 /video-10s1080P( 16:9 & 9:16 )$1.22 /video-10s1080P( 4:3 & 3:4 )$1.22 /video-10s1080P( 1:1 )$1.22 /videoVidu 2.0 Image to Video-4s360P$0.09 /video-4s720P$0.18 /video-4s1080P$0.27 /video-8s720P$0.27 /videoVidu 2.0 Reference to Video-4s360P$0.09 /video-4s720P$0.18 /videoVidu 2.0 Start End to Video-4s360P$0.09 /video-4s720P$0.18 /video-4s1080P$0.27 /video-8s720P$0.27 /videoVidu Q1 Image to Video-5s1080P$0.36 /videoVidu Q1 Reference to Video-5s1080P$0.36 /videoVidu Q1 Start End to Video-5s1080P$0.36 /videoVidu Q1 Text to Videogeneral style5s1080P$0.36 /videoanime style5s1080P$0.36 /videoVIDU Q2 Pro Fast Image to Video-5s720P$0.0713 /video-5s1080P$0.1430 /videoVIDU Q2 Pro Fast Start-End Frame to Video-5s720P$0.0713 /video-5s1080P$0.1430 /videoVIDU Q2 Pro Image to Video-5s540P$0.1472 /video-5s720P$0.2454 /video-5s1080P$0.5135 /videoVIDU Q2 Pro Multi-frame to Video-5s540P$0.0357 /video-5s720P$0.0224 /video-5s1080P$0.1785 /videoVIDU Q2 Pro Start-End Frame to Video-5s540P$0.1472 /video-5s720P$0.2454 /video-5s1080P$0.5135 /videoVIDU Q2 Reference Image to Video-5s540p$0.1562 /video-5s720p$0.2008 /video-5s1080p$0.5132 /videoVIDU Q2 Template to Video20credit-$0.0900 /video40credit-$0.1800 /video60credit-$0.2700 /video80credit-$0.3600 /video100credit-$0.4500 /video120credit-$0.5400 /video180credit-$0.8100 /video8credit-$0.0360 /video30credit-$0.1350 /video45credit-$0.2025 /video48credit-$0.2160 /video78credit-$0.3510 /video140credit-$0.6300 /video240credit-$1.0800 /video460credit-$2.0700 /video110credit-$0.4950 /video145credit-$0.6525 /video200credit-$0.9000 /videoVIDU Q2 Text to Video-5s540P$0.0802 /video-5s720P$0.1562 /video-5s1080P$0.2677 /videoVIDU Q2 Turbo Image to Video-5s540P$0.0624 /video-5s720P$0.2141 /video-5s1080P$0.3347 /videoVIDU Q2 Turbo Multi-frame to Video-5s540P$0.0179 /video-5s720P$0.0357 /video-5s1080P$0.1117 /videoVIDU Q2 Turbo Start-End Frame to Video-5s540P$0.0624 /video-5s720P$0.2141 /video-5s1080P$0.3347 /videoVidu Q3 Pro Image to VideoOff-Peak-540P$0.0313 /sPeak-540P$0.0625 /sOff-Peak-720P$0.067 /sPeak-720P$0.1339 /sOff-Peak-1080P$0.0714 /sPeak-1080P$0.1429 /sVidu Q3 Pro Image-to-VideoOFFPEAK-540p$0.0313 /sPEAK-540p$0.0625 /sOFFPEAK-720p$0.0670 /sPEAK-720p$0.1339 /sOFFPEAK-1080p$0.0714 /sPEAK-1080p$0.1429 /sVidu Q3 Pro Start-End-to-VideoPEAK-1080p$0.1429 /sOFFPEAK-1080p$0.0714 /sPEAK-720p$0.1339 /sOFFPEAK-720p$0.0670 /sPEAK-540p$0.0625 /sOFFPEAK-540p$0.0313 /sVidu Q3 Pro Text to VideoOff-Peak-540P$0.0313 /sPeak-540P$0.0625 /sOff-Peak-720P$0.067 /sPeak-720P$0.1339 /sOff-Peak-1080P$0.0714 /sPeak-1080P$0.1429 /sVidu Q3 Pro Text-to-VideoPEAK-1080p$0.1429 /sOFFPEAK-1080p$0.0714 /sPEAK-720p$0.1339 /sOFFPEAK-720p$0.0670 /sPEAK-540p$0.0625 /sOFFPEAK-540p$0.0313 /sVidu Q3 Turbo Image-to-VideoOFFPEAK-540p$0.0179 /sPEAK-540p$0.0357 /sOFFPEAK-720p$0.0268 /sPEAK-720p$0.0536 /sOFFPEAK-1080p$0.0357 /sPEAK-1080p$0.0714 /sVidu Q3 Turbo Start-End-to-VideoPEAK-1080p$0.0714 /sOFFPEAK-1080p$0.0357 /sPEAK-720p$0.0536 /sOFFPEAK-720p$0.0268 /sPEAK-540p$0.0357 /sOFFPEAK-540p$0.0179 /sVidu Q3 Turbo Text-to-VideoPEAK-1080p$0.0714 /sOFFPEAK-1080p$0.0357 /sPEAK-720p$0.0536 /sOFFPEAK-720p$0.0268 /sPEAK-540p$0.0357 /sOFFPEAK-540p$0.0179 /sWan 2.1 Image to VideoLoRA / Fast-720P$0.2250 /videoLoRA / Fast-480P$0.1250 /videoLoRA-720P$0.3000 /videoLoRA-480P$0.2000 /videoFast-720P$0.2250 /videoFast-480P$0.1250 /videoStandard-720P$0.3000 /videoStandard-480P$0.2000 /videoWan 2.1 Text to VideoLoRA / Fast-720P$0.2250 /videoLoRA / Fast-480P$0.1250 /videoLoRA-720P$0.3000 /videoLoRA-480P$0.2000 /videoFast-720P$0.2250 /videoFast-480P$0.1250 /videoStandard-720P$0.3000 /videoStandard-480P$0.2000 /videoWan 2.2 Image to VideoNo LoRA5s1080P$0.4000 /videoNo LoRA5s480P$0.0900 /videoNo LoRA5s720P$0.2700 /videoNo LoRA8s480P$0.2880 /videoNo LoRA8s720P$0.4320 /videoLoRA5s480P$0.1800 /videoLoRA5s720P$0.3150 /videoLoRA8s480P$0.2880 /videoLoRA8s720P$0.5040 /videoWan 2.2 Text to VideoNo LoRA5s832*480$0.0900 /videoNo LoRA5s1280*720$0.2700 /videoNo LoRA5s1920*1080$0.4000 /videoNo LoRA8s832*480$0.2880 /videoNo LoRA8s1280*720$0.4320 /videoLoRA5s832*480$0.1800 /videoLoRA5s1280*720$0.3150 /videoLoRA8s832*480$0.2880 /videoLoRA8s1280*720$0.5040 /videoWan 2.5 Image to Video-5s480P$0.25 /video-10s480P$0.50 /video-5s720P$0.50 /video-10s720P$1.00 /video-5s1080P$0.75 /video-10s1080P$1.50 /videoWan 2.5 Image to Video Preview-5s1080P$0.7500 /video-5s480P$0.2500 /video-10s480P$0.5000 /video-5s720P$0.5000 /video-10s720P$1.0000 /video-10s1080P$1.5000 /videoWan 2.5 Text to Video-5s480P$0.25 /video-10s480P$0.50 /video-5s720P$0.50 /video-10s720P$1.00 /video-5s1080P$0.75 /video-10s1080P$1.50 /videoWan 2.5 Text to Video Preview-5s832*480$0.2500 /video-10s832*480$0.5000 /video-5s1280*720$0.5000 /video-10s1280*720$1.0000 /video-10s1920*1080$1.5000 /video-5s1920*1080$0.7500 /videoWan 2.6 Image to Video-5s720P$0.50 /video-10s720P$1.00 /video-15s720P$1.50 /video-5s1080P$0.75 /video-10s1080P$1.50 /video-15s1080P$2.25 /videoWan 2.6 Image To Video-5s720P$0.5000 /video-10s720P$1.0000 /video-15s720P$1.5000 /video-5s1080P$0.7500 /video-10s1080P$1.5000 /video-15s1080P$2.2500 /videoWan 2.6 Reference to Video-5s720P$0.50 /video-10s720P$1.00 /video-5s1080P$0.75 /video-10s1080P$1.50 /videoWan 2.6 Text to Video-5s720P$0.50 /video-10s720P$1.00 /video-15s720P$1.50 /video-5s1080P$0.75 /video-10s1080P$1.50 /video-15s1080P$2.25 /videoWan 2.6 Text to Video-5s1280*720$0.5000 /video-10s1280*720$1.0000 /video-15s1280*720$1.5000 /video-5s1920*1080$0.7500 /video-10s1920*1080$1.5000 /video-15s1920*1080$2.2500 /videoWan 2.6 Video Reference-5s1280*720$0.5000 /video-10s1280*720$1.0000 /video-5s1920*1080$0.7500 /video-10s1920*1080$1.5000 /videoAPI NameModelStepsPricingImage to VideoSVD-XT20$0.024 /videoSVD20$0.0134 /videoAudioAPI NameModePricingFish Audio Text to Speech-$15 /1M charactersFish Audio Voice Cloning-$0.1 /voiceGLM Audio to Text-$0.0210 /MtGLM Text to Speech-$0.2800 /10K charactersGLM Voice Clone-$0.8300 /voiceMiniMax Speech 2.8 HD Async Text-to-Speech-$100.0000 /1M charactersMiniMax Speech 2.8 HD Sync Text-to-Speech-$100.0000 /1M charactersMiniMax Speech 2.8 Turbo Async Text-to-Speech-$60.0000 /1M charactersMiniMax Speech 2.8 Turbo Sync Text-to-Speech-$60.0000 /1M charactersMiniMax speech-02-hdT2A / T2A Async$80 /1M charactersMiniMax speech-02-turboT2A / T2A Async$48 /1M charactersMiniMax speech-2.5-hd-previewT2A / T2A Async$80 /1M charactersMiniMax speech-2.5-turbo-previewT2A / T2A Async$48 /1M charactersMiniMax speech-2.6-hdT2A / T2A Async$100 /1M charactersMiniMax speech-2.6-turboT2A / T2A Async$60 /1M charactersMiniMax Voice-Cloning-$2.4 /voiceText to Speech-$15 /1M charactersReady to build smarter? Start today.Get started with Novita AI and unlock the power of affordable, reliable, and scalable AI inference for your applications.Get StartedBook a Demo --- Model LibraryExplore models ready for production — deploy in seconds.FeaturedAll ModelsLLMServerlessImageAudioVideoEmbeddingRerankerVisionProviders:AllHotDeepseek V3.2$0.269/MtInput$0.1345/MtCache Read$0.4/MtOutput163840Context65536Max OutputLLMServerlessNewGLM-5-Turbo$1.2/MtInput$0.24/MtCache Read$4/MtOutput202800Context131072Max OutputLLMServerlessNewMiniMax M2.7$0.3/MtInput$0.06/MtCache Read$1.2/MtOutput204800Context131072Max OutputLLMServerlessNewMiniMax M2.5-highspeed$0.6/MtInput$0.03/MtCache Read$2.4/MtOutput204800Context131100Max OutputLLMServerlessNewQwen3.5-397B-A17B$0.6/MtInput$3.6/MtOutput262144Context65536Max OutputLLMServerlessNewMiniMax M2.5$0.3/MtInput$0.03/MtCache Read$1.2/MtOutput204800Context131100Max OutputLLMServerlessNewGLM-5$1/MtInput$0.2/MtCache Read$3.2/MtOutput202800Context131072Max OutputLLMServerlessNewQwen3 Coder Next$0.2/MtInput$1.5/MtOutput262144Context65536Max OutputLLMServerlessNewDeepSeek-OCR 2$0.03/MtInput$0.03/MtOutput8192Context8192Max OutputLLMServerlessNewKimi K2.5$0.6/MtInput$0.1/MtCache Read$3/MtOutput262144Context262144Max OutputLLMServerlessMinimax M2.1$0.3/MtInput$0.03/MtCache Read$1.2/MtOutput204800Context131072Max OutputLLMServerlessGLM-4.7$0.6/MtInput$0.11/MtCache Read$2.2/MtOutput204800Context131072Max OutputLLMServerlessNewMXiaomiMiMo/MiMo-V2-Flash$0.1/MtInput$0.02/MtCache Read$0.3/MtOutput262144Context32000Max OutputLLMServerlessNewAutoGLM-Phone-9B-Multilingual$0.035/MtInput$0.138/MtOutput65536Context65536Max OutputLLMServerlessNewKimi K2 Thinking$0.6/MtInput$0.15/MtCache Read$2.5/MtOutput262144Context262144Max OutputLLMServerlessMiniMax-M2$0.3/MtInput$0.03/MtCache Read$1.2/MtOutput204800Context131072Max OutputLLMServerlessDeepseek V3.2 Exp$0.27/MtInput$0.41/MtOutput163840Context65536Max OutputLLMServerlessNewQwen3 VL 235B A22B Thinking$0.98/MtInput$3.95/MtOutput131072Context32768Max OutputLLMServerlessNewGLM 4.6V$0.3/MtInput$0.055/MtCache Read$0.9/MtOutput131072Context32768Max OutputLLMServerlessNewGLM 4.6$0.55/MtInput$0.11/MtCache Read$2.2/MtOutput204800Context131072Max OutputLLMServerlessKat Coder Pro$0.3/MtInput$0.06/MtCache Read$1.2/MtOutput256000Context128000Max OutputLLMServerlessDeepSeek-OCR$0.03/MtInput$0.03/MtOutput8192Context8192Max OutputLLMNewDeepseek V3.1 Terminus$0.27/MtInput$0.135/MtCache Read$1/MtOutput131072Context32768Max OutputLLMServerlessNewQwen3 VL 235B A22B Instruct$0.3/MtInput$1.5/MtOutput131072Context32768Max OutputLLMServerlessQwen3 Max$2.11/MtInput$8.45/MtOutput262144Context65536Max OutputPartnerLLMServerlessDeepSeek V3.1$0.27/MtInput$0.135/MtCache Read$1/MtOutput131072Context32768Max OutputLLMServerlessKimi K2 0905$0.6/MtInput$2.5/MtOutput262144Context262144Max OutputLLMServerlessQwen3 Coder 480B A35B Instruct$0.3/MtInput$1.3/MtOutput262144Context65536Max OutputLLMServerlessOpenAI GPT OSS 120B$0.05/MtInput$0.25/MtOutput131072Context32768Max OutputLLMServerlessKimi K2 Instruct$0.57/MtInput$2.3/MtOutput131072Context131072Max OutputLLMServerlessHotDeepSeek V3 0324$0.27/MtInput$0.135/MtCache Read$1.12/MtOutput163840Context163840Max OutputLLMServerlessQwen3 235B A22b Thinking 2507$0.3/MtInput$3/MtOutput131072Context32768Max OutputLLMServerlessQwen3 235B A22B Instruct 2507$0.09/MtInput$0.58/MtOutput131072Context16384Max OutputLLMServerlessDedicated EndpointEnterprise-Grade Infrastructure for AIFor enterprises that require higher performance, tailored SLAs, or private hosting for custom modelsCustom pricingGuaranteed uptime & latencyUnlimited scaleDedicated clustersGet Enterprise-Grade Endpoint --- Let your AI Agents run for realAgent Sandbox is a secure, fast, and programmable runtime designed for real-world AI Agent execution.Get StartedPricingMillisecond-Level StartupSandbox instances launch in under 200 ms on average — optimized for high-frequency and concurrent workloads.Secure IsolationEach task runs in a fully isolated environment with system-level separation to prevent data leakage and unauthorized access.High ConcurrencyRun thousands of sandbox instances in parallel with consistently low latency — ideal for real-time, high-throughput workloads.AGENTSANDBOXSpeed, Security, andConcurrencyLaunch fast. Stay isolated. Handle thousands of tasksin parallel — all in real-time.Try NowSandbox CapabilitiesCode ExecutionRun Python, JavaScript, C++, and other languages in a secure sandbox.Network AccessAllow agents to access external APIs and online data as needed.Browser UseAutomate web navigation, form filling, and content scraping.Computer UseControl full GUI environments: input text, scroll pages, take screenshots, and more.Session PersistencePause and resume long-running agent tasks without losing progress.Visual OutputEnable agents to interact with GUI and stream task execution via VNC.Flexible, Usage-Based PricingPay only for what you use — billed per second based on vCPU and memory. No plans, no lock-ins, no waste. Learn moreSample ConfigurationUsage ExampleEstimated Cost1 vCPU + 512 MiB RAMShort task running for 5 minutes~$0.00322 vCPU + 1 GiB RAMBatch code execution for 1 hour~$0.07638 vCPU + 8 GiB RAMMulti-agent parallel processing (1 hr)~$0.3283Ready to Explore Agent Sandbox?Explore Agent Sandbox and start building with our step-by-step guides.QuickstartCreate your first agent sandbox now.Learn moreTemplateDefine once, reuse everywhere.Learn moreFilesystemEach Sandbox has a completely isolated filesystem.Learn moreEnterpriseFor advanced workloads requiring large memory, higher vCPU, or custom deployment locations — we offerFlexible resource configurationsDeployment to your preferred cloud or private infrastructureBook a Call