Decide better.
Pay less than you think.

A synthesis is any time Forge combines AI responses into a structured output — clicking "Run Synthesis", "Best Answer →", or "Analyze →" in Excel. Getting the eight individual AI perspectives is always free and unlimited. You only use a synthesis credit when Forge produces the combined, decision-ready result.

No. Forge uses its own backend access to all eight providers — Claude, ChatGPT, Gemini, Mistral, DeepSeek, Perplexity, Grok, and Meta AI. Your Forge subscription is all you need. If you already have your own subscriptions and want to use your own accounts via the browser extension, you can do that too — but it is not required.

White-label means your exported reports carry your name and company instead of Forge's. It's for analysts, consultants, and advisors who deliver work to clients and need the output to look like theirs — not like it came from an AI tool. Takes two minutes to set up on first export. Available on Work Like a Pro and above.

They are a structured intake that gives eight AI analysts the domain knowledge they need to interpret your data correctly — not just compute it. Without context, AI applies generic patterns to specialist problems. With all seven questions answered, Forge produces analysis that's accurate to your specific domain, data source, column definitions, and decision objective. Read the full explanation on the Why Excel page →

Yes. Upgrade, downgrade, or cancel at any time from your Profile page. If you cancel, your current plan remains active until the end of the billing period. No cancellation fees and no questions asked.

Forge is operated by Xencore Global GmbH under Swiss jurisdiction. Your prompts and AI responses are sent to the AI providers you select, under their respective privacy policies. We do not train models on your data and do not retain your analysis content after your session ends. Full details on the Privacy page →

We've built a proprietary intelligence layer underneath Forge that improves with every synthesis made on the platform. We don't share the technical details of how it works — but what it means for you is that the synthesis quality you experience today is better than what early users received, and it continues to improve. It's one of the fundamental differences between Forge and simply using ChatGPT or Claude directly. They answer questions. Forge accumulates decision intelligence.

One question reaches eight models simultaneously — you see where they agree and where they diverge, which tells you something no single AI can tell you. High consensus means a robust answer. High divergence means a genuinely contested question you should read carefully. The synthesis layer then builds on that signal to produce one calibrated output. No single AI can show you its own blind spots.