Fast mode runs the same Claude Opus 5 model at up to 2.5x the output speed, for double the price: $10 per million input tokens and $50 per million output. It is not a smaller or distilled model — the intelligence is identical, only the throughput and the bill change.
The trade
| Standard | Fast mode | |
|---|---|---|
| Input (per 1M) | $5 | $10 |
| Output (per 1M) | $25 | $50 |
| Output speed | Baseline | Up to 2.5x |
| Model quality | Identical | Identical |
| Rate limits | Standard pool | Separate pool |
Because the rate limits are separate, fast mode can also be a capacity valve rather than purely a latency choice.
Where you can use it
Fast mode is a research preview on the Claude API only, including Managed Agents. It is not available on Amazon Bedrock, Google Cloud, or Microsoft Foundry, and it does not work with the Batch API or Priority Tier.
In Claude Code, the /fast toggle switches it on directly — the same model, faster output, at the premium rate.
When it is worth doubling the price
- Interactive products where a user is watching a response appear and latency is the experience.
- Long agentic runs where wall-clock time is the constraint rather than token cost.
- Capacity pressure, since fast mode draws on its own rate-limit pool.
It is rarely worth it for background jobs, batch work, or anything a user is not waiting on. If cost matters more than speed, the opposite lever — the Batch API at half price — is the one to reach for.
One caveat when switching
Switching speed invalidates the prompt cache. If you fall back from fast mode to standard on a rate limit, you also pay to rebuild your cached prefix — worth accounting for before you build automatic fallback logic.
More on Claude Opus 5
Start with the complete Claude Opus 5 guide for the overview, or go deeper:
Ready to go deeper?
Read the full Claude Opus 5 guideFrequently Asked Questions
Is fast mode a different model?
No. It is the same Claude Opus 5 running at higher output throughput — up to 2.5x the tokens per second. Quality is identical; only speed and price change.
How much does Claude Opus 5 fast mode cost?
$10 per million input tokens and $50 per million output — exactly double the standard rate. It also has its own rate-limit pool, separate from standard Opus 5 usage.
Can I use fast mode on Amazon Bedrock or Google Cloud?
No. Fast mode is available on the Claude API only, including Managed Agents. It is not offered on Amazon Bedrock, Google Cloud, or Microsoft Foundry, and it is incompatible with the Batch API and Priority Tier.
