No message caps · No cooldowns
Claude-level AI for the refactorsthat don't fit in a message limit.
// agent.ts
const res = await draftify
.chat({ model: "fable-5.5" })
You
Refactor the auth module and keep every test green.
$ claude --base-url api.draftify.site
✓ connected · Opus 5.5-level
Chapter 01 — The wall
Every other AI stops you mid-thought.
You've reached your message limit.
Upgrade or try again later.
Usage
100%
Resets in 3h 59m
Chapter 02 — The way through
Draftify doesn't.
Tokens this session
0
Limit: none
Remaining
∞
Chapter 03 — The models
Frontier-class models. None of the waiting.
Chapter 04
Now read the fine print.
00 / 04 — Prologue
04 — Draftify vs. a typical AI subscription
We redlined the fine print.
¶ 01 · Message limits
Typical subscription: Message caps that reset every few hours, and cooldowns at peak times.
Draftify: No message caps on paid plans. No cooldowns.
¶ 02 · Models
Typical subscription: Fable 5.5, Opus 5.5 and Sonnet 5.5-level models.
Draftify: All of them, in one workspace.
¶ 03 · Developer API
Typical subscription: A separate developer console, billed separately.
Draftify: Prepaid API credits. No subscription required.
¶ 04 · Reasoning
Typical subscription: Thinking that's hidden or out of your hands.
Draftify: Collapsible reasoning, with timing for every step.
¶ 05 · Your tools
Typical subscription: Manual setup for every tool you use.
Draftify: One endpoint for Claude Code, Cursor, Codex and more.
¶ 06 · Data retention
Typical subscription: Zero data retention, but only on business plans.
Draftify: Zero retention on every plan.
05 — Plugs into what you already use
- Claude Code✳
- Cursor✳
- Codex CLI✳
- Kilo Code✳
- Opencode✳
- Trae✳
- Droid✳
- Grok CLI✳
06 — Developer API
Pay for tokens.
Not for waiting.
Pay only for the tokens you use. No message caps, no cooldowns, and no monthly subscription.
Read the docs| Model | Input / m | Output / m |
|---|---|---|
| Fable 5.5-level | 2.5€ | 6€ |
| Fable 5-level | 2€ | 5€ |
| Opus 5.5-level | 1€ | 3€ |
| Opus 5-level | 0.50€ | 1.50€ |
| Sonnet 5.5-level | 0.2€ | 0.74€ |
m = million tokens.
07 — Plans
Pick your capacity.
Starter
€40/ mo
€33.33/mo equivalent · save €80
- Unlimited Sonnet 5-level
- Fable 5 and 5.5-level, with a usage allowance
- Standard support
Pro
€90/ mo
€75.00/mo equivalent · save €180
- Everything in Starter
- Unlimited Opus 5.5-level
- Priority support
Scale
Most popular€130/ mo
€108.33/mo equivalent · save €260
- Everything in Pro
- Unlimited Fable 5 and 5.5-level
- 24/7 priority support
Which models can I use?
Sonnet 5.5, Opus 5.5, Fable 5 and Fable 5.5-level models. Which ones are unlimited depends on your plan; the plans above spell it out.
Is there really no message limit?
On paid plans, every model marked unlimited has no message caps and no cooldowns. The free plan has a daily request limit so you can try Draftify out, and Fable on Starter and Pro comes with a usage allowance.
Do you train on my code?
No. Your prompts and workspace context are only used to generate your response, and never to train models. The details are in our Privacy Policy.
What's the difference between Pro and Scale?
Pro (€90/mo) adds unlimited Opus 5.5-level to everything in Starter. Scale (€130/mo) adds unlimited Fable 5 and Fable 5.5-level on top of that. The API is billed separately, pay-as-you-go, on any plan.
Can I use the API in my own product?
Yes. Use it in commercial products, internal tools or automations. You own what you build, and you pay only for the tokens you use.
Does Draftify support Skills and tool use?
Yes. Skills (SKILL.md files) and function calling both work, so you can build agents the same way you would with the Claude API.