Get Started
Announcement · Mar 8, 2026 · Release Notes
v2026.3.8

QVerisBot 2026.3.8: Memory Upgrade, Model Switching & 60+ Bug Fixes

The biggest QVerisBot update yet — persistent memory, seamless model switching, smarter capability routing, and over 60 resolved issues.

60+
bugs fixed
longer memory context
4
new models available

What's New in v2026.3.8

NEW

🧠 Persistent Memory Upgrade

QVerisBot's memory system has been completely rebuilt. Context window is now 3× longer, with smarter summarization that preserves key facts across long conversations.

  • Context window expanded from 32K to 96K tokens
  • Smart summarization preserves entities and decisions
  • Memory now persists across sessions (opt-in)
  • New /memory command to view and edit stored context
QVerisBot — Terminal
# New memory commands: /memory show # View current memory /memory clear # Clear session memory /memory pin "fact" # Pin a fact permanently
NEW

🔀 Model Switching

Switch between four frontier models mid-conversation without losing context. Each model brings different strengths — pick the right one for each task.

Claude 3.7 Sonnet

Best for: reasoning & code

GPT-4o

Best for: general tasks

Gemini 2.0 Pro

Best for: multimodal

DeepSeek R2

Best for: cost efficiency

QVerisBot — Terminal
# Switch model in chat: /model claude # Switch to Claude 3.7 Sonnet /model gpt4o # Switch to GPT-4o /model gemini # Switch to Gemini 2.0 Pro /model deepseek # Switch to DeepSeek R2

Model switching preserves full conversation history and QVeris tool access.

IMPROVED

🔀 Smarter Capability Routing

The routing engine behind QVerisBot's tool calls has been upgraded with better intent parsing and provider selection logic.

  • Intent detection accuracy improved by 23%
  • Provider selection now weights recent success rates
  • Parallel tool calls for multi-step queries
  • Automatic retry with fallback provider on failure
  • Tool call latency reduced by avg 180ms

Before v2026.3.8

71%
Intent accuracy

After v2026.3.8

94%
Intent accuracy
IMPROVED

📋 Response Formatting

Responses are now better structured with automatic markdown formatting, cleaner table rendering, and improved code block handling.

  • Auto-detects when to use tables vs prose
  • Code blocks now include language detection
  • Financial data always rendered as formatted tables
  • Long responses include automatic section headers
  • Copy button on all code blocks and tables

60+ Bugs Fixed

Every reported issue from the last 90 days — resolved.

Memory & Context 15 fixes

  • Fixed memory duplication on session reload
  • Resolved context overflow causing truncation
  • Fixed pinned facts being lost after 24h
  • Memory export now includes full conversation tree
  • Fixed /memory clear not persisting across tabs
  • + 10 more fixes

Tool Calls & Routing 18 fixes

  • Fixed race condition in parallel tool calls
  • Resolved auth token expiry during long sessions
  • Fixed JSON parsing errors on malformed API responses
  • Tool call retry now respects user timeout settings
  • Fixed duplicate tool calls on network retry
  • + 13 more fixes

Model & Response 14 fixes

  • Fixed model switching losing system prompt
  • Resolved streaming interruptions on slow connections
  • Fixed code block rendering in markdown responses
  • Model preference now saved per conversation
  • Fixed response cutting off at exactly 2048 tokens
  • + 9 more fixes

UI & Experience 13 fixes

  • Fixed dark mode flicker on page load
  • Resolved mobile keyboard pushing chat input off screen
  • Fixed copy button not working in Firefox
  • Conversation list now loads incrementally
  • Fixed search not finding older conversations
  • + 8 more fixes

Recent Release History

v2026.3.8
Mar 8, 2026
Memory upgrade · Model switching · 60+ bug fixes
CURRENT
v2026.2.15
Feb 15, 2026
QVeris network expanded to 10,000+ capabilities
v2026.1.20
Jan 20, 2026
QVerisBot public launch · 500+ providers · MCP support
v2025.12.1
Dec 1, 2025
Private beta launch · Core routing engine
QVerisBot v2026.3.8 Highlights

How to Get the Update

🌐

Web App

The update is live automatically at qveris.ai/qverisbot. No action required — just open the app.

Open QVerisBot →

💻

API

If you're calling QVerisBot via API, new features are available immediately. Update your model parameter to access model switching.

View API Docs →

🔌

MCP Server

Update your QVeris MCP Server to the latest version for improved routing and memory in Cursor and Claude Code.

npx @qverisai/mcp@latest

MCP Docs →

Frequently Asked Questions

What's new in QVerisBot v2026.3.8?
Three major upgrades: persistent memory with a 3× larger context window (96K tokens), mid-conversation model switching across 4 frontier models, and smarter capability routing with 23% better intent detection. Plus 60+ bug fixes across memory, tool calls, model switching, and UI.
How does QVerisBot's memory upgrade work?
The new memory system expands context from 32K to 96K tokens with smart summarization that preserves key entities and decisions. Memory can now persist across sessions (opt-in), and new /memory commands let you view, edit, pin, and clear stored context at any time.
Can I switch AI models in QVerisBot?
Yes. Use /model claude, /model gpt4o, /model gemini, or /model deepseek to switch between four frontier models mid-conversation without losing conversation history or QVeris tool access.
How do I update to the latest version?
The web app at qveris.ai/qverisbot updates automatically. For API users, new features are available immediately. For MCP Server, run npx @qverisai/mcp@latest to update to the latest version.
How many bugs were fixed in v2026.3.8?
Over 60 bugs were resolved across four categories: memory and context (15 fixes), tool calls and routing (18 fixes), model and response handling (14 fixes), and UI and experience (13 fixes). All reported issues from the last 90 days were addressed.