AI BriefingAI Gateway adds Opus 4.7 fast mode
Key point
Vercel AI Gateway now offers fast mode for Claude Opus 4.7 as a research preview.
Details
Fast mode for Claude Opus 4.7 is now available on Vercel AI Gateway as a research preview. This mode boosts output token generation speed by about 2.5x while maintaining Opus 4.7's level of intelligence, according to the explanation.
To enable it, specify providerOptions.anthropic.speed: "fast" for the anthropic/claude-opus-4.7 model. As an example, adding speed: "fast" to a streamText call was presented.
Fast mode is also available via Claude Code via AI Gateway. To use it, you need to set the CLAUDE_CODE_ENABLE_OPUS_4_7_FAST_MODE=1 and CLAUDE_CODE_SKIP_FAST_MODE_ORG_CHECK=1 environment variables in your shell configuration or ~/.claude/settings.json.
Pricing is 6x the standard Opus rate.
- Standard: input $5 / 1M tokens, output $25 / 1M tokens
- Fast mode: input $30 / 1M tokens, output $150 / 1M tokens
Standard price multipliers, such as those for prompt caching, are calculated additionally on top of this rate.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
Claude Code Fails to Load AGENTS.md When Telemetry Is Disabled, With WorkaroundClaude Code Update: Memory Warnings, MCP Timeout Controls, and Stability Improvements