ClaudeDevs
@ClaudeDevs
Fast mode now defaults to Opus 4.7 in Claude Code. Try it out today with /fast https://t.co/i0roMdEgEg
331retweets6.5klikes
View on XAnthropic Makes Opus 4.7 the Default for Claude Code Fast Mode
Anthropic· Updated
Anthropic updated Claude Code to use Opus 4.7 as the default model for its high-speed Fast mode configuration. The update delivers identical reasoning quality at 2.5x the response speed by prioritizing low latency over cost efficiency for interactive developer workflows.
Anthropic updated Claude Code to use Opus 4.7 as the default for Fast mode. This configuration provides identical reasoning quality to standard inference but at 2.5x the speed. It follows the OpenRouter Fast mode for Opus 4.7 launch on third-party platforms.
- Model
- Opus 4.7
- Speed increase
- 2.5x
- Pricing (input)
- $30 per million tokens
- Pricing (output)
- $150 per million tokens
- Availability
- Claude Code v2.1.139 or later
- Billing
- Usage credits only
This update addresses the latency bottleneck in agentic engineering, where model thinking time can disrupt developer flow. By prioritizing speed over cost, Anthropic enables rapid iteration. This mirrors OpenAI's Fast mode for Codex, following a pattern seen in OpenAI's Codex Remote Connections for managing interactive agent sessions.
You can toggle the feature using the /fast command in version 2.1.139 or later. Fast mode is billed at $30 per million input tokens and $150 per million output tokens, drawing from usage credits rather than subscription limits. It is best reserved for time-sensitive interactive work rather than long autonomous tasks.
Still wondering? A few quick answers below.
Fast mode is a high-speed configuration for Anthropic's Claude Opus models within the Claude Code CLI and VS Code extension. It uses a specific API setup to deliver identical reasoning quality and capabilities as standard Opus but at approximately 2.5x the response speed, prioritizing low latency for interactive developer tasks.
Fast mode is billed at a higher rate than standard Opus, costing $30 per million input tokens and $150 per million output tokens. These charges are deducted from your account's usage credits rather than your plan's included rate limits, meaning you must have usage credits enabled in your Anthropic Console settings to use it.
Fast mode is exclusively available for the Claude Opus model family, specifically Opus 4.6 and Opus 4.7. It is not supported on other Anthropic models like Sonnet or Haiku. As of May 2026, Opus 4.7 is the default model for Fast mode, though users can manually override this to use Opus 4.6.
You can toggle Fast mode on or off by typing the /fast command in the Claude Code terminal or VS Code extension. When active, a lightning bolt icon appears next to the prompt. You must be running Claude Code version 2.1.139 or later to use Opus 4.7 as the default Fast mode model.
Fast mode and effort levels both impact speed but through different mechanisms. Fast mode uses a faster API configuration to reduce latency without changing model quality. In contrast, lowering the effort level reduces the model's internal thinking time, which results in faster responses but may lead to lower quality on complex reasoning tasks.
Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →
