Skip to main content

​
How Pricing Works

  1. Sign up and get $0.50 free credits (no credit card needed)
  2. Use any model through the same API endpoint
  3. Pay per token after credits run out
  4. No monthly fees. No commitments. No hidden costs.

​
Model Pricing

All prices per 1 million tokens.
ModelInputOutput~Cost / 1K reqs
For the current canonical pricing table, use:
  • GET /v1/models
  • GET /v1/credits/pricing
Common active prices today:
ModelInputOutput~Cost / 1K reqs
gemma-4-31b$0.19$0.54$0.20
gpt-oss-120b$0.20$0.81$0.26
qwen-3-32b$0.39$0.81$0.36
minimax-m2.5$0.41$1.62$0.53
minimax-m2.7$0.41$1.62$0.53
qwen-3.6-plus$0.44$2.63$0.75
qwen-3-coder$0.68$2.16$0.77
llama-3.3-70b$1.19$1.19$0.83
deepseek-v3$0.81$2.30$0.86
gemini-2.5-flash$0.41$3.38$0.88
deepseek-r1$0.68$2.90$0.92
kimi-k2.5$0.68$3.78$1.09
gemini-3-flash$0.68$4.05$1.15
Typical cost per request assumes 500 input + 200 output tokens. Your actual cost depends on message length. Sorted cheapest first.

​
Cost Examples

Use CaseTypical TokensModelCost
Quick chat (1 message)200 in / 100 outgemma-4-31b~$0.00009
Code review (1 file)2,000 in / 500 outqwen-3-coder~$0.002
Blog article500 in / 2,000 outdeepseek-v3~$0.005
Data extraction (1 page)1,000 in / 200 outqwen-3-32b~$0.0004
RAG query (large context)10,000 in / 500 outgemini-2.5-flash~$0.006

​
Free Credits

  • $0.50 on signup (no credit card required)
  • Enough for approximately 500 to 3,000 requests depending on model
  • Credits never expire

​
Adding Credits

PackageAmount
Starter$5
Growth$20
Pro$100
Scale$500
Auto top-up supported (set threshold + amount). Pay via Stripe.

​
Prompt Caching

Cached input tokens cost 10% of normal price (90% discount). Caching works automatically with compatible providers. Average cache hit rate across platform: 48%. Learn more about prompt caching

​
Next Steps