ClaudeCode No Flicker Mode Changes Everything!
Claude Code released a 'no flicker mode' that eliminates screen flashing and memory bloat issues in terminal usage. The update includes mouse support and uses an alternate screen buffer to provide a smoother, more stable coding experience.
Summary
Claude Code has introduced a significant update called 'no flicker mode' that addresses one of the most persistent problems with terminal-based AI coding tools. The flicker issue occurred because Claude Code is built with React rendering to terminals, which weren't designed for interactive UIs, causing the entire terminal buffer to redraw with every update. The Anthropic team had previously reduced flickering by 85% through a rendering system rewrite, but about a third of sessions still experienced flickers, especially in VS Code Terminal, TMux, and iTerm 2. The new no flicker mode can be activated with the environment variable 'CLAUDE_CODE_NO_FLICKER=1' and switches to an alternate screen buffer similar to Vim or htop. This approach keeps the input box fixed at the bottom, only renders visible messages rather than entire conversation history, and maintains flat memory usage even in long sessions. Unexpectedly, the update also includes full mouse support in the terminal, allowing users to click to move cursor position, expand tool outputs, scroll with mouse wheel, and click URLs directly. Users can disable mouse capture while keeping flicker-free rendering using 'CLAUDE_CODE_DISABLE_MOUSE=1'. The feature is marked as a research preview and requires Claude Code v2.188 or later. There are some trade-offs including conversation history living in the alternate buffer rather than terminal scrollback, and potential clipboard copy issues over SSH due to OSC 52 escape sequences. Anthropic has been working on multiple fixes including upstream patches to VS Code's terminal and TMux for synchronized output support, making this part of a comprehensive approach to improving terminal-based AI coding tools.
Key Insights
- Claude Code's flicker problem occurs because it's built with React rendering to terminals that weren't designed for interactive UIs, causing entire terminal buffer redraws with every update
- Anthropic previously reduced flickering by 85% through a rendering system rewrite, but about a third of sessions still experienced at least one flicker
- No flicker mode uses an alternate screen buffer and only renders messages visible on screen rather than the entire conversation history, keeping memory usage flat even in long sessions
- The update unexpectedly includes full mouse support in the terminal, allowing users to click to move cursor, expand outputs, scroll, and click URLs directly
- Anthropic has been pushing upstream fixes including patches accepted into VS Code's terminal and TMux to add synchronized output support, eliminating flicker at the source
Topics
Transcript
[0:00] Clawed Code, no flicker mode, changes everything. Clawed Code just fixed one of the most annoying problems in the terminal. No more screen flashing, no more random jumps to the top of your screen. And yes, they added mouse support in a terminal. This is one of those updates that sounds small but feels massive the second you try it. Let me break it all down. So, if you've been using Claude code in the terminal, you know the flicker problem. You're running a long task, Claude [music] is doing its thing, and the whole screen just flashes. Oops. He draws over and over again. It's distracting. It's annoying [0:30] and for a lot of people, especially if you're…
Full transcript available for MurmurCast members
Sign Up to AccessMore from Julian Goldie SEO
NEW Nvidia Autonomous AI is WILD!! 🤯
Nvidia announced Nemo Clo, a new autonomous AI agent system that operates independently without continuous prompting. Powered by Nemotron 3 Ultra (a 550 billion parameter model), the system is five times faster and cheaper than previous versions, with OpenShell providing secure sandboxed execution.
Laguna XS 2.1: New FREE + Opensource Local AI!
Julian reviews Laguna XS 2.1, a new free open-source local AI coding model from Poolside that performs comparably to Qwen 3.6 and outperforms Claude Haiku on benchmarks. He demonstrates its practical capabilities by building landing pages and functional apps, highlighting its speed, offline functionality, and multiple deployment options through local setup, Claude Code, or OpenRouter's free API.
How to Run Hermes FREE Forever!
The video demonstrates how to run the Hermes AI agent for free using Gemma 4, a local open-source model from Google, with significant speed improvements through MLX optimization. The setup works on Apple Silicon Macs or via free APIs on Open Router, enabling autonomous agents to work offline and privately without subscription costs.
This NEW Chinese AI is INSANE! (FREE + Open Source!)
Long Cap 2.0 is a new open-source Chinese AI model from a food delivery app company that offers 1 million tokens of free context memory, beats GPT-4.5 on SWE bench pro benchmarks, and uses efficient parameter activation to reduce computational overhead while maintaining high performance.
Claude Code is now FREE: Here’s how…
Google's new Gemma 4 model running on Ollama is 90% faster on Apple Silicon, enabling free Claude Code usage locally without token costs. The setup requires three simple steps: downloading Ollama, Gemma 4, and installing into Claude Code, with alternatives available via OpenRouter API for non-Mac users.