We’ve always believed that great content fuels great go-to-market strategies. Whether you’re in marketing, sales, or enablement, your teams need the power to create assets that stand out, connect, and convert. That’s why we’ve made a massive upgrade to Dreamwriter’s engine and we’re excited to show you what it can do.

Let’s break down what’s new and why it matters for every GTM team using Dreamwriter.

New Brains Behind the Magic: Our Multi-Large Language Model (LLM) Stack

We’re no longer a one-model shop. Previously, Dreamwriter ran almost entirely on GPT-4o and 4o Mini. It worked well, but “well” isn’t our bar. 

So we got to work testing, comparing, and optimizing across the most advanced models on the market. The result?  We’ve moved to a smarter, faster, and more specialized approach- a custom multi-LLM stack. Think of it as assembling the dream team for content creation. 

Here's the new lineup:

• Claude 3.7 Sonnet Claude now handles the heavy lifting — writing outlines, selecting page formats, and generating your slide/page content. Claude is structured, thoughtful, and brilliant at creating content that flows logically while sounding human (and on-brand).

• Gemini Flash Need to rewrite content quickly? Dreamy now taps Gemini Flash; a model that’s ultra-fast and excellent at rewording while preserving meaning. Whether you're simplifying, rephrasing, or iterating, it's done in a flash.

• DeepL For translations, we’ve integrated DeepL-  widely regarded as the best AI translation tool available today. Your content doesn’t just translate; it adapts, making you sound fluent in every market you enter.

This isn’t just a swap of engines; it’s a full-stack elevation of how Dreamwriter thinks, writes, and adapts. Each LLM was chosen based on how well it performs in the specific part of the content workflow it's powering. 

Each model plays to its strengths, and together, they make Dreamwriter smarter, faster, and more reliable for GTM teams who can’t afford to slow down.

Why This Matters for GTM Teams

Marketing and sales teams don’t have time to babysit AI. You want content that’s accurate, engaging, and personalized, without needing five rounds of editing. This upgrade is all about getting you closer to the “final draft” quality in a single click.

We're also excited to unveil our new RAG vector database (Retrieval-Augmented Generation). 

In simple terms: Dreamwriter now has a memory.

This means the platform can reference your product data, past collateral, tone of voice guidelines, and more, all in real-time, while generating new content. 

No more starting from scratch. No more manual tweaks to align with your pitch. Dreamwriter just gets it and builds on what you’ve already created.

It’s like giving your AI a memory, and that memory happens to know your product, your pitch, and your tone of voice.

Coming Next Week: Choose Your Visual-Text Ratio

Different audiences. Different formats. Different vibes.

We know every team and every campaign has a different need when it comes to design and format. Some need image-rich slides that pop on a Zoom call. Others want detailed text-driven docs to share with decision-makers.

So, starting next week, you’ll be able to choose from three content styles when generating collateral in Dreamwriter:

• Text Heavy, Image Light: Ideal for datasheets, one-pagers, or technical enablement.

• Balanced Text and Images: Perfect for pitch decks or sales presentations.

• Heavy Images, Less Text: Designed for visual storytelling, social slides, or event content.

This gives you even more control and helps you align content format with your audience, channel, and goals.

Always Improving, Never Settling

This multi-LLM stack is just the beginning. Our promise to you is simple: we’ll always be testing and adopting the best models available to keep Dreamwriter at the cutting edge. 

Our goal is to ensure that your GTM teams never settle for generic content again.

So go ahead and put Dreamwriter’s new brainpower to the test.

👉 Check out the new Dreamwriter and see what smarter content creation looks like.