Claude Code Fast Mode: What /fast Does and What It Costs
Fast mode makes Claude Opus answer up to 2.5x faster in Claude Code, at a higher per-token price. Which models support it, how to turn it on with /fast, why it bills from usage credits on subscription plans, and when it's worth paying for.
Fast mode is a setting that makes Claude Opus respond up to 2.5x faster in Claude Code. It's the same model with the same quality — Anthropic runs it with a configuration that prioritises speed over cost, and charges more per token for it.
Details as of October 2026. Fast mode is a research preview, so pricing and availability may change; see Anthropic's fast mode docs.
Which models
Fast mode works on Opus 5.5, Opus 5 and Opus 4.8. It isn't available on Sonnet or Haiku. If you turn it on while using a model that doesn't support it, Claude Code switches you to Opus.
How to turn it on
/fast
Press Space to toggle, Enter to confirm. You'll see "Fast mode ON" and a small ↯ icon next to the prompt. Run /fast again to turn it off (you stay on Opus).
You can also set "fastMode": true in your user settings. By default, fast mode stays on across sessions once you've enabled it.
What it costs
| Model | Fast mode input | Fast mode output |
|---|---|---|
| Opus 5.5 | $8 per million tokens | $40 per million tokens |
| Opus 5 / Opus 4.8 | $10 per million tokens | $50 per million tokens |
For comparison, standard Opus 5.5 is $4 / $20 — so fast mode doubles the price.
On Pro, Max, Team and Enterprise plans, fast mode is paid from usage credits only. It doesn't come out of your plan's included usage, even if you have plenty left. You need usage credits turned on (Settings → Usage on claude.ai, or run /usage-credits). Team and Enterprise owners must enable fast mode for the organisation first.
On a Claude Console (API) account, it's billed per token with the rest of your usage.
A cost trap to avoid
The first time you turn fast mode on in a conversation, you pay the fast-mode price for re-reading the whole conversation so far without the cache discount. Deep into a long session, that's expensive. Turn it on at the start of a session, not halfway through. (Prompt Caching Explained)
When it's worth it
Fast mode suits work where you're sitting and waiting:
- Rapid back-and-forth on a UI.
- Live debugging with a deadline.
- Pairing sessions where latency breaks your flow.
It's poor value for:
- Long autonomous tasks you'll check on later.
- CI pipelines and batch jobs.
- Anything where cost matters more than a few seconds.
For speed without extra cost, try lowering the effort level on simple tasks instead — less thinking, faster answers. (Effort Levels and Ultrathink)
If it turns itself off
Fast mode has its own rate limits. When you hit them, it falls back to normal speed automatically (the ↯ turns grey) and comes back after a cooldown. If your usage credits run out, Claude Code keeps working at normal speed and price and tells you fast mode was disabled.
The summary
/fastmakes Opus up to 2.5x faster at double the price.- Subscription users pay from usage credits, not plan usage.
- Turn it on at the start of a session, and only when you're actively waiting.
EasySpawn runs Claude Code on a persistent server, so the long tasks you don't need fast mode for can run at normal speed while you do something else. See how it works or join the waitlist.
Related: Claude Code Usage Limits · Claude Pro vs Max vs API · Claude API Pricing Explained · How to Keep Claude Code Costs Down
Keep reading
Is Claude Code Free? What You Actually Need to Use It
The Claude Code software is free to install, but using it needs a paid Claude plan (Pro, Max, Team or Enterprise) or an API account billed per token. What each option costs, what the cheapest realistic setup is, and what the "free Claude Code" guides online are really doing.
Claude Opus vs Sonnet vs Haiku: Which Model Should You Use?
Anthropic's Claude models come in tiers — Opus, Sonnet and Haiku, plus Fable at the top. What each is good at, what they cost on the API, and a simple rule for choosing for coding, chat and features inside your own app.