What's New in Merciful: Tiers, ETAs, and Themes

September 2026 · by the Merciful team

When we launched Merciful, the premise was simple: AI chat without friction, walls, or surveillance. Over the past few months, we've been upgrading our entire frontend and runner architecture to make the chat faster, more reliable, and more flexible.

Here is a direct look at the new features live in your chat interface today.

1. Three Distinct Model Tiers

Different problems need different tools. Rather than locking everyone into a single compromise model, conversations can now leverage three purpose-built model tiers:

2. Real-Time Response Progress & ETAs

No more wondering if a backend dropped your message. The chat interface now displays an active progress bar with server-reported status and estimated time-to-completion while the model processes your query.

3. Email Verification & Account Matrix

You can still chat completely anonymously with no signup required. But if you want to save room history and unlock larger token budgets, verifying an email address gives you an immediate upgrade at zero cost:

Account Level Daily Tokens Max Rollover Model Access
Anonymous (No signup) 10,000 30,000 Standard & Economy
Verified Free (Email verified) 50,000 150,000 Advanced (thread starters) + Standard & Economy
Pro ($5/mo) 500,000 1,500,000 Unlimited Advanced model + Priority routing

4. Fresh UI Themes & Account Controls

In your settings modal, you can now toggle between three curated themes:

You also have direct controls to set model preferences, custom system prompts, and manage notification settings.

5. The $5/Month Pro Plan

We introduced Merciful Pro for power users who want continuous, heavy access to our highest-tier models. At $5/month, it provides half a million tokens every day and bypasses message caps on the Advanced model.

Because we run our own GPU infrastructure, every Pro subscription directly subsidizes compute costs for our free community.

Curious about the technical and economic choices behind these changes? We've published a companion deep dive: Why How We Run Changed: Dedicated Compute and Sustainable AI.

Tell Us What You Think

We take user feedback very seriously. If an answer looks off, if latency spikes, or if you have an idea for what would make Merciful more helpful for your day-to-day work, click the feedback button in chat. We read every submission.

— The Merciful Team

← Back to Blog How We Run →