Transparent Wholesale Pricing
Prepaid credit, billed per token as the response streams. Enjoy Claude models at 70% lower cost with zero monthly minimums.
Deposit funds into your dedicated on-chain address. Every request deducts exact token costs in real time.
Wholesale rates, zero retail markup
We aggregate volume across developers to deliver rates you cannot get alone. Prices are per million tokens.
| Model | Official Price (In / Out) | InfinityRouter Rate (In) | InfinityRouter Rate (Out) | You Save |
|---|---|---|---|---|
Claude Opus 5claude-opus-5 | $15.00 / $75.00 | $4.50 | $22.50 | Save 70% |
Claude Sonnet 5claude-sonnet-5 | $3.00 / $15.00 | $0.90 | $4.50 | Save 70% |
Gemini 7 Flashgemini-7-flash | $0.15 / $0.60 | $0.075 | $0.30 | Save 50% |
GPT-5gpt-5 | $5.00 / $15.00 | $2.50 | $10.00 | Wholesale |
DeepSeek R2deepseek-r2 | $0.80 / $3.20 | $0.55 | $2.19 | Wholesale |
Claude 3.5 Haikuclaude-haiku-4 | $0.80 / $4.00 | $0.24 | $1.20 | Save 70% |
See how much you will save
1M Input Tokens · 200k Output Tokens
- Official Anthropic$6.00
- InfinityRouter$1.80
10M Input Tokens · 2M Output Tokens
- Official Anthropic$60.00
- InfinityRouter$18.00
100M Input Tokens · 20M Output Tokens
- Official Anthropic$600.00
- InfinityRouter$180.00
How billing works
1. Temporary Deposit Hold
When a request starts, we temporarily hold a small deposit from your balance for the expected answer so your request never gets cut off.
2. Live Word Counting
As words stream back to you, tokens are counted live. If an AI provider fails before returning words, you pay $0.
3. Instant Balance Settle
When the response finishes or if you cancel it early, the unused hold is instantly refunded and you only pay for what you received.
Add Balance with Crypto (USDC & USDT)
Generate your personal deposit address in the dashboard. Payments are detected in seconds and added to your balance right after network confirmation.
- Supported NetworksBase (Ethereum L2), Tron
- Accepted CoinsUSDC, USDT
- Minimum Deposit$5.00
- Speed1 to 2 minutes
- Billing PrecisionExact per token (no rounding up)
- Active Models0 models
Programmatic Pricing API
from openai import OpenAI
client = OpenAI(
base_url="https://infinityrouter.qd.je/v1",
api_key="inf_live_...",
)
# Fetch current catalog with capabilities and token prices
models = client.models.list()
for model in models.data:
pricing = getattr(model, "pricing", {})
input_rate = pricing.get("input_per_million_tokens", "N/A")
output_rate = pricing.get("output_per_million_tokens", "N/A")
print(f"Model: {model.id} -> In: ${input_rate}/1M | Out: ${output_rate}/1M")Frequently Asked Questions
How do you provide Claude models 70% cheaper?
We aggregate high-volume compute commitments across global tier-1 infrastructure suppliers and pass wholesale token pricing directly to you without retail markups.
Do you charge for failed or dropped requests?
No. If a request is rejected before dispatch, fails mid-generation, or you disconnect your client, you are billed only for the tokens delivered to your stream before disconnect.
Does prepaid credit expire?
Never. Your balance remains available until consumed. There are no monthly maintenance fees or seat charges.
Which payment methods are supported?
We accept stablecoin deposits (USDC and USDT) on Base and Tron networks with automated on-chain confirmation and instant balance credits.
Data handling differs by plan. See the privacy policy