AI/ML API

❄️ Winter Holiday gift grom AI/ML API: Up to 5% extra credits, ends Jan 15! - Stocking → +3% bonus credits (great for exploring new models) - Gift Box → +5% bonus credits (ideal for growing teams) - Sleigh Full → 5% cashback in credits (made for heavy usage) Last chance, ends January 15. Claim your holiday gift here: https://lnkd.in/dxvWR5br

Wan 2.6 is here!!! 🎥 ✨ The new video model synchronizes audio and video, creates clips from multiple cameras, and adds sound design. What it can do: generate up to 15 seconds in 1080p, multi-shot videos with multiple scenes in a single prompt, audio support, and character consistency via references. The model supports audio-video synchronization, multi-camera generation, sound effects, and more. In image generation, it introduces precise control over lighting and optics, plus the ability to work with multiple references at once, crucial for advertising and brands. The AI itself cuts scenes into shots: it changes behavior and lines while preserving the face and voice. Lip-syncing is accurate, and the audio is perfectly synchronized. Longer stories, sharper shots with Wan 2.6 on AI/ML API: https://lnkd.in/gZZKM3Dj

Join us in 2 hours (at 16:00 UTC) for our Live Session: AI Art Director for a Global Marketplace. Divo Solutions' CEO Daria Kamyshina and CTO Victor Kamyshin-Pisarevskii will walk through a real world use case of an AI Art Director built for a global marketplace. Join us on Discord here: https://lnkd.in/d-B7kR2Y

Google DeepMind introduces Gemini 3 Flash, frontier intelligence built for speed: - ~3× faster than Gemini 2.5 Pro - Costs a fraction of Pro (≈1/4) with comparable or better performance - Strong at coding, tool calling, and agentic workflows - More token-efficient → lower latency and lower cost Try Gemini 3 Flash now on AI/ML API and build apps in seconds! https://lnkd.in/g6jKF9ws

We’re continuously evolving to make AI more accessible, flexible, and intuitive. Today, we’re excited to share a series of updates designed to simplify your experience and expand what you can create. Here’s what’s new: 🔹 Single Billing Model We’re moving toward a single, flexible Pay As You Go model, with no monthly commitments. You only pay for what you use, and both fiat and crypto payments are supported. New users start directly on Pay As You Go. Existing users will keep their current plan for the next 30 days before automatically switching to Pay As You Go. 🔹 Expanded Model Access in Playground We’ve added all model categories to the Playground, including music, speech, embeddings, and OCR. Already available without limits: chat, code, image, video, and music models. By the end of the week, all categories will be fully accessible in the UI, complete with their configuration parameters. 🔹 Smarter, More Intuitive Playground Manual parameter configuration is now a thing of the past. Model parameters are automatically set based on your selection, ensuring stable and high quality outputs right away. You still have full control to adjust settings anytime you like. Try out the updates and build with more freedom: https://aimlapi.com/app

How do you save 80% of an Art Director's time with AI? Join us on Dec 18 at 16:00 UTC for a live session with Divo Solutions' CEO Daria Kamyshina and CTO Victor Kamyshin-Pisarevskii in AI/ML API Discord server. We're diving deep into a real world case: building AI Art Director with AI/ML API. Watch, ask questions, and see how we created a system that automates design QA for a major global marketplace. See you there: https://lnkd.in/dhBe7jRn

🔥 Inworld AI TTS-1 & TTS-1-Max are here! Inworld TTS-1 Max is the new leader on the Artificial Analysis Speech Arena Leaderboard, surpassing MiniMax's Speech-02 series and OpenAI's TTS-1 series. With support for 12 languages and instant voice cloning, you get SOTA quality for every use case. We've also added a free ready made n8n workflow to put this power to use. It listens for new Gmail emails, uses GPT-5 to generate a natural short summary, and then converts it into a lifelike voice message using Inworld TTS-1-Max for instant delivery via a Telegram bot. No servers or hardcoding are required. n8n workflow: https://lnkd.in/gXFcDqiQ Experience the #1 ranked speech AI:  https://lnkd.in/gUKgSEeY https://lnkd.in/g2EQzziE

GPT-5.2 is live on AI/ML API on day 0 ⚡ GPT-5.2 Thinking has a human expert level. The model that sets a new standard for professional work: - World-class long-context reasoning. - Spreadsheet creation, analysis, and formatting on a new level. - A breakthrough in preparing presentations and complex artifacts. GPT-5.2 Instant is built for everyday tasks and learning. Just as warm and conversational, but even clearer: - Explains clearly, highlighting key points. - Better step-by-step instructions and walkthroughs. - Enhanced technical writing and translation. - A reliable assistant for study and career. GPT-5.2 Pro is the smartest and most reliable model for difficult questions: - Unmatched performance in programming and complex domains. - The best tool for accelerating scientific research and discovery. A frontier model surpasses human experts in ~70% of job tasks and is built for the long intellectual haul.

MythoMax-L2-13B is back on our platform! A long time favorite for creative power, rich roleplay, and storytelling. The model has impressive function calling abilities. We had MythoMax-L2-13B and GPT-5 Chat act like a golden retriever. Which response do you like more?