⚡ Quick Summary
- ✓ Universal credits — use with any provider
- ✓ Traditional MT — charged per character (~1-2 credits/char)
- ✓ AI Translation — charged per token (2-90 credits/translation)
- ✓ Mix & match — use different providers for different tasks
- ✓ Monthly reset — credits refresh each billing cycle
How Credits Work
Credits are Tolkie's universal currency for machine translation. Instead of managing separate accounts with Google, DeepL, or OpenAI, you get a simple credit balance that works with all providers.
What You Get
- • Free plan: 10,000 credits/month
- • Early Adopter: 100,000 credits/month
- • Add-on: +25,000 credits for $1/mo
Monthly Reset
Credits reset at the start of each billing cycle. Unused credits don't roll over — use them or lose them!
Traditional Machine Translation
Fast, reliable engines — charged per character
| Provider | Rate | Tier | Languages |
|---|---|---|---|
| Azure Translator | 0.8 credits/char | Budget | 100+ |
| AWS Translate | 1.2 credits/char | Standard | 75+ |
| Google Translate | 1.6 credits/char | Standard | 130+ |
| DeepL | 1.8 credits/char | Premium | 30+ |
Example Calculation
Translating "Welcome to our application" (27 characters):
- • Azure: 27 × 0.8 = 22 credits
- • Google: 27 × 1.6 = 43 credits
- • DeepL: 27 × 1.8 = 49 credits
AI Translation
Higher quality translations — charged per token
What are tokens?
Tokens are how AI models read text — roughly 4 characters or 0.75 words per token. AI translation has both input tokens (your text + instructions) and output tokens (the translation).
| Model | Input Rate | Output Rate | Tier | Per Translation |
|---|---|---|---|---|
| GPT-5 nano Fast & cheap | 0.004 | 0.032 | Budget | ~2 credits |
| GPT-5 mini Better quality | 0.02 | 0.16 | Budget | ~10 credits |
| Claude Haiku 4.5 Great balance | 0.08 | 0.4 | Standard | ~30 credits |
| GPT-5 High quality | 0.1 | 0.8 | Standard | ~42 credits |
| GPT-4.1 Reliable | 0.16 | 0.64 | Standard | ~54 credits |
| Claude Sonnet 4.5 Best quality | 0.24 | 1.2 | Premium | ~90 credits |
* "Per Translation" assumes ~300 input tokens (text + system prompt) and ~15 output tokens (translated string).
Example Calculation
Translating a short string with GPT-5 nano (300 input + 15 output tokens):
# GPT-5 nano
credits = (300 × 0.004) + (15 × 0.032)
credits = 1.2 + 0.48
credits = 1.68 (rounded to 2)
With 10K free credits, you can translate ~5,000 strings with GPT-5 nano.
📊 Credits Calculator
Estimated Credits Usage
When to Use Which?
🌍 Traditional MT
Best for:
- ✓ Short UI strings (buttons, labels)
- ✓ High volume, simple content
- ✓ When speed matters most
- ✓ Budget-conscious translation
🤖 AI Translation
Best for:
- ✓ Marketing copy & descriptions
- ✓ Nuanced or creative text
- ✓ Quality-critical translations
- ✓ When traditional MT sounds "robotic"
❓ Frequently Asked Questions
What happens if I run out of credits?
Translation requests will be blocked until your credits reset (next billing cycle) or you purchase a credits add-on. Your existing translations remain accessible — you just can't create new ones.
Why is AI translation sometimes cheaper than traditional MT?
Budget AI models like GPT-5 nano have very low token rates. For short strings, the token count is minimal, making it cheaper than per-character pricing. However, AI has a fixed overhead (system prompt tokens), so for very short strings or bulk operations, traditional MT might be more efficient.
Do unused credits roll over to the next month?
No, credits reset each billing cycle. This keeps pricing simple and predictable. If you consistently need more credits, consider adding a credits add-on ($1/mo for 25K credits).
What's included in the "input tokens" for AI?
Input tokens include: your source text and system prompt (translation instructions). Typically ~250-350 tokens for a standard translation request.
Are there any hidden fees?
No hidden fees. The credit rates shown include our margin. What you see is what you pay. Your subscription covers platform access, and credits cover translation API costs.
Ready to get started?
Join the waitlist and get 100,000 credits/month with the Early Adopter plan.
Join waitlist