# Adaline is now generally available. Giving $1MM in API credits. Canonical URL: https://www.adaline.ai/blog/adaline-is-generally-available-giving-1mm-in-api-credits LLM text URL: https://www.adaline.ai/blog/adaline-is-generally-available-giving-1mm-in-api-credits/llms.txt Published: 2025-06-17T00:00:00.000Z Modified: 2025-06-17T08:17:56.809Z Author: Arsh Shah Dilbagi Category: Product Visibility: public Reading time: 15 min Topics: Product, Adaline, AI agent observability, agent evals, self-improving agents ## Summary After nearly a year of building with some incredible customers and over 100K developers, Adaline is now open to everyone for iteration, evaluation, deployment, and monitor with even more coming soon. ## Article # Introducing Adaline: The Platform for Building with Intelligence Today, we’re opening the doors. After nearly a year of obsessive iteration, relentless engineering, and deeply listening to our early users, Adaline is now **generally available**. Anyone can **[get started](/get-started)**. Anyone can start building. And we can’t wait to see what you’ll inspire. Video: https://a-us.storyblok.com/f/1023026/x/b6bc15899e/adaline-demo-video.mp4 # A New Interface for Intelligence The arrival of LLMs isn’t just a new wave of technology. It’s a shift in how we interface with machines. The prevailing narrative says it's painful and hard to build with LLMs because the output is stochastic. We disagree. At its core, a large language model is still just a function — input goes in, output comes out. The real change is in the **modality** of input. We’ve gone from code that can be linted and compiled to natural language that inspires. You’ll never write a novel in VSCode. And you’ll never debug code in Google Docs. So where do your instructions to LLMs (prompts) go? Where do your teams design the future? # Built for the Builders of the Future We designed Adaline for the builders — for the engineers, researchers, PMs, designers, and domain experts inventing new workflows and shipping AI-native products. **Because we believe that the next generation of software won’t just be coded — it will be instructed.** And the people instructing it won’t just be developers. It’ll be the entire team. Adaline is where that team works together. Adaline gives you one unified platform to iterate, evaluate, deploy, and monitor your prompts and intelligent applications — collaboratively, transparently, and with rigor. That’s how fast-moving teams build reliable AI applications. That’s how the future ships. # Trusted by the Best Some of the most iconic companies in the world have already built with Adaline — from Salesforce, DoorDash, Discord, HubSpot, McKinsey, and many more. They’ve trusted us to power mission-critical LLM applications, shape internal tools, and speed up go-to-market velocity. They didn’t just need another devtool. They needed a new kind of platform — one built for a new era. # A Moment Worth Marking We’re at an inflection point that happens once in a century. LLMs are the wheel. We’ve just begun to craft the first horse carriages. The roads, the engines, the Teslas, and the F1s? They’re yet to come. But to build that future, we need solid foundations. We need shared infrastructure, shared language, and shared tools. That’s what Adaline aims to be. A platform that doesn't just keep up with the future — It helps you shape it. Video: https://a-us.storyblok.com/f/1023026/x/5aae7e3f1a/what-will-you-inspire.mp4 **To celebrate, we’re giving $1,000,000 in Adaline API credits to the first 100 eligible team workspaces. **(_Details in the FAQ below._) # We’re Just Getting Started Today is exciting. It’s humbling. And, truthfully, a little terrifying. But we believe deeply in what we’ve built. We’ve poured our minds, craft, and care into every pixel, every interaction, every function. It’s all designed to help you build faster, smarter, and with more confidence. So here it is. Adaline is live. Ready when you are. **[Get Started 🚀](/get-started)** Let’s make the future feel obvious. # FAQ ### What does Adaline actually do? Adaline is the end-to-end platform for product and engineering teams building AI applications. Teams use Adaline to iterate, evaluate, deploy, and monitor their AI prompts. **Iterate** Build and refine your prompts within a powerful collaborative editor and playground. - **Dynamic Prompting**: Use variables in your prompt to simulate all use cases. - **LLM Playground**: Run your prompt across top LLM providers in a safe sandbox. - **Playground History**: View your prompt changes across playground runs with rollback options. **Evaluate** Run, test, and validate your prompt performance with real test cases at scale. - **Evaluators**: Choose evaluators like LLM-as-a-judge, text matcher, JavaScript, cost, token usage, etc. - **Linked Dataset**: Store and organize thousands of evaluation test cases in datasets. - **Analytics**: View detailed evaluation reports with scores, cost, token usage, latency, etc. **Deploy** Ship your prompts to production with confidence. - **Version Control**: View and manage your prompt changes across versioned deployments. - **Deployment Environments**: Deploy to isolated environments for safe CI/CD releases. - **Cross-environment Control**: Promote and rollback deployments between environments **Monitor** Monitor your AI app using telemetry and continuous evaluations. - **Observability**: View, search, and filter on real-time and historical traces and spans. - **Continuous Evaluation**: Monitor prompt performance using continuous evaluations running on live telemetry data. - **Analytics**: View curated time-series charts of latency, token usage, cost, evaluation scores, and more. ### How does Adaline differ from other prompt tools or LLM playgrounds? Most prompt tooling today is either: - **Too simple** (just a basic playground without evaluation or monitoring) - **Too technical** (ML tools not designed for cross-functional product teams) - **Too fragmented** (you need 4–5 different tools to build a real workflow) Adaline is different because: - It’s **one interface** for your entire lifecycle—from prompt design to post-deploy QA - It includes **advanced evaluation at scale**, such as human-like feedback loops, LLM-as-a-judge, and customizable rubrics - It enables **non-ML teammates** to participate in shipping quality LLM-powered features - It comes with **real-time observability** and deployment flexibility out of the box In short, it’s a modern product stack for prompt engineering. ### Can I use this if I already have an internal prompt testing process? Yes—and we’ll likely replace or enhance it. Adaline is often used by teams who are currently managing: - Prompt versions in Notion, code comments, or spreadsheets - Testing via ad hoc Python scripts - Evaluation with inconsistent human feedback - Deployments hard-coded into RAG/Agent frameworks Adaline centralizes and professionalizes this process, giving your team **confidence to ship faster** with **repeatable results**. ### What models and integrations does Adaline support? Adaline natively supports over 300+ LLMs, including ones from: - **OpenAI** - **Anthropic** - **Gemini, Bedrock, Azure, TogetherAI, OpenRoute, Custom Providers **and more You can also connect your own API keys or bring your **custom endpoints** if needed. Deployment integrations are flexible—most teams use Adaline to support: - **Agent-based architectures (w/ MCPs, Tools, etc.)** - **RAG systems (Retriever-Augmented Generation)** - **Custom LLM-based workflows inside their product** ### Can I invite my team to use the credits together? Yes. Adaline is built for **team collaboration**—your workspace allows you to: - Share prompt templates - Collaborate on evaluations - Track version changes and notes - Deploy and monitor together - Manage API usage centrally The $10K credit is tied to the workspace—not just the individual account—so your whole team can benefit. ### Who is eligible for the $1MM in API credits? We are offering $10,000 in API credits to the first 100 qualifying team workspaces. To be a qualified team workspace, you must: - Use a company email address to get started. - Work at a company with at least 15 employees. - Be in a senior Product or Engineering role, particularly those who are actively building AI applications with LLMs. - Subscribe to the yearly Grow plan and be out of the 7-day trial window. Our team manually reviews eligibility within 48 hours of the first payment of the yearly Grow plan to ensure fairness and alignment with the program's intent. You should expect to hear during the week of July 14, 2025. ### Is there a deadline to claim the $10K credit? Yes. The $1M pool is **first-come, first-qualified. **Once 100 qualified accounts have been approved, the credit program closes. There is no waiting list. ### What happens if I'm eligible and get the credits? You’ll get a confirmation email, and your credits will be provisioned within 24–48 hours after eligibility is verified. You can start using Adaline immediately in the meantime—your workspace will automatically reflect the balance once approved. We will also send you: - A quick onboarding walkthrough - Best practice templates - Link to a dedicated Slack/Teams channel - Meeting link to schedule a pair prompting and concierge onboarding session ### Where do the API credits apply? The $10K API credits can be applied towards your "usage" subscription (not the seat-based subscription) as shown in your Stripe billing portal. You can access the Stripe billing dashboard from Settings > Billing > Billing Portal. The usage-based subscription applies to overages towards Logs, Deployments, Continuous Evaluations, Dynamic Workers, and more. ### What happens if I get started and I’m not eligible? If you don’t meet the eligibility criteria (e.g., wrong role, company too small, personal email), you can still use Adaline with all its features. You will still get the 7-day free trial on the Grow plan. ### Is this a trial or do I keep the credits if I qualify? This is **not a time-limited trial.** The $10K credit is a full usage balance that expires after **3 months** after the date it is granted. Once approved, you can consume it at your own pace—no forced acceleration, no billing surprises.